Federalism by wuyyok


The Optimum Size and Shape of
  with an Application to Crime
           Alexander Tabarrok
         Department of Economics
         George Mason University

    Why Federalism? or What is the
     Optimal Size and Structure of
•   The Pure Economic Theory
     •   The technology of public goods production drives the
         organization of production.

     •   Illustrative question. In most countries, the central
         government provides military defense while local
         governments provide city parks. Why? What
         characteristics of these goods explains the nature of the

     •   Before we answer that question, however, we need to ask
         why have government at all? Or how should the
         production of goods be divided between market and
                             Public Goods
•   A “public good” is a non-excludable and non–
    rivalrous good.
•   Non-excludability means that people who don’t
    pay can still consume – free riders. The market will
    underproduce a good if free riding is common.
•   Non-rivalry means that your consumption of the
    good does not impede my consumption. We both
    can benefit from consuming the same national
    defense but cannot both benefit from consuming
    the same apple.

         • “He who receives an idea from me, receives
           instruction himself without lessening mine; as he who
           lights his taper at mine, receives light without
           darkening me.” Thomas Jefferson

•   Note that non-rivalry implies that even if it were
    possible to exclude it would not be efficient to do
•   Classic example of a public good: National
• Government can overcome the problem of
  free riders by forcing people to pay thus
  increasing the provision of public goods.
    • Problem of forced riders.
    • Just because government can increase the
      provision of public goods doesn’t mean that it will.
      The Economic Theory of Federalism is normative
      not positive.
• Given that government should provide
  some public goods we can turn to the
  question which governments should
  produce which public goods?
•   Some public goods like basic science are very “pure” in the sense
    that everyone in the world could potentially benefit from their
    production. Other public goods, like flood control, are more location
    or group specific. In addition, some public goods may become
    congested. National defense protects 200 million people just as
    easily as 100 million but a park is non-rivalrous only so long as it
    isn’t crowded.
  From the Technology of Production to the
         Organization of Production
• If “transaction costs” were zero then in theory any type of
  public good could be produced optimally by any stucture
  of government.
• E.g. In theory, national defense could be provided by a
  combination of states that bargained with one another to
  reach an optimal plan.
• In practice, bargaining is expensive, information is not
  free and it is difficult to transmit from the periphery to the
  center, and political institutions do not perfectly align
  incentives and public interests.
       • Bargaining will not eliminate the free rider problem among states so
         the public good of national defense will be under provided. E.g.
         Articles of Confederation. NATO.
•   At the other extreme, local parks could be provided by a national
    government which carefully gathered information on the demand for parks
    by location and allocated taxes to coincide with benefits.
         • When central government tries to produce local public goods it could underprovide (I’m
           not paying for a project in someone else’s district) or overprovide (pork) but is unlikely
           to provide at optimal levels.
•   E.g. Consider a city with three neighborhoods.
         • The central urban area is densely populated, few homes have yards.
         • The near suburbs have homes with small yards.
         • The far suburbs have homes with large yards.
•   How will the demand for the public good of parks differ among these three
•   Who should decide how many parks to build?

•   If the question is decided by the city as whole some groups are bound to be
    disappointed. If the near and far suburbs form a voting bloc we are likely to
    see few parks, which satisfies the far suburbs (because they have low
    demand for parks) but upsets those in the central urban area.
•   If the central urban area dominates we are likely to see lots of parks which
    leads those in the far suburb to feel they are over-taxed.
•   Or we could have a medium number of parks which satisfies those in the
    near suburbs but no other groups.

•   These considerations motivate the matching rule for government provision.
      The Matching Rule for Government
• Match the scope of the government to the scope of the
  public good. i.e. Assign finance and control to the
  smallest unit of government that matches the scope of
  the public good.
• National defense should be provided by the national
  government. Support for basic research should be
  provided nationally, or even more appropriately
   – The WHO, for example, helps to finance and coordinate the
     production of influenza vaccines. Without world finance we are
     likely to get free riders and too little vaccine production (Similarly
     for basic research, asteroid detection and destruction, world
• Local public goods should be matched to local
• Returning to our example of the city with three
  neighborhoods the solution is to match the government
  to the scope of the public good – create three areas,
  “special districts,” and let them choose independently.

• Note that heterogeneity of preferences is a key concept.
  Homogeneous groups are likely to have similar demands
  for public goods and will be more easily able to provide
  those goods collectively and centrally.

• “Many activities that may be quite rationally collectivized
  in Sweden, a country with a relatively homogeneous
  population, should be privately organized in India,
  Switzerland, or the United States.”
          Buchanan and Tullock, The Calculus of Consent.
The Matching Rule is About the Optimal
   Size and Shape of Governments

    What is the key public good/resource in this picture?
    What area should the relevant government cover?
    What area(s) does it cover?
                               Special Districts
•   Special districts are independent government authorities that typically have control
    over one subject area. There are 35,052 special districts in the United States (2002).

•   Common subject areas for special districts:
                     Drainage and flood control                   Soil and water conservation
                     Sewage                                       Parks
                     Water supply                                 Fire protection

•   These sorts of activities will often have natural boundaries that do not coincide with
    political boundaries (as in the Detroit-Windsor river example). As a result, most
    special districts do not coincide with county boundaries, either crossing over or
    residing only partially within them.

•   Special districts allow for better matching of public good scope to government scope.
    On the other hand, there may be cost savings (economies of scope in production and
    better voter monitoring) from combining functions within a single government entity.
          • Same issue as in private markets. Do you want to buy both your groceries and clothes
            at Wal-Mart or do you want to buy at two different stores which may be more expensive
            but which allows for better matching to your particular tastes?

•   Special districts have tripled in number since 1952, compared to no increase in the
    number of counties and virtually no increase in the number of towns, suggesting that
    more government activities have spillover effects requring better matching or that the
    cost of having multiple governments has fallen.
                          Special Districts (con’t)
•   Special districts do not have to follow the one-person, one-vote rule.
          • The Supreme Court has held, for example, that a water storage district's
             limited purpose and the disproportionate effect of its activities on landowners
             meant that limiting the right to vote on special district matters to landowners
             or weighting votes by land value does not violate the equal protection clause
             of the 14th Amendment.
                – Salyer Land Company v. Tulare Lake Basin Water Storage District (1973)

           • Homeowner Associations, including condominiums, also often weight votes
             (and taxes!) according to property value.

           • Similar to matching principle – match those who pay with those who receive

•   A virtue of special districts is that they can grow or shrink over time.
     –   In 1789 the United States had 13 states and four million people. Not that much different
         from contemporary Switerzland which has 26 cantons and approximately 7 million people.
     –   If the number of states had grown as fast as the number of people or if we in the United
         States had about the same amount of decentralization as do the contemporary Swiss we
         would today have about 1000 states.
     –   If that sounds extreme why is 50 the magic number? And why is 50 the magic number when
         the population is 150 million as when it is 300 million?
                        Tiebout Competition
•   The pure economic theory emphasizes how the technology of public good
    production influences the optimal structure of government organization.
•   The pure ET assumes efficient and benevolent government.
•   When we recognize that men are not angels, government organization
    becomes even more important for reasons that go beyond the benefits of
•   Charles Tiebout (1956) showed that under ideal conditions voting with
    one’s feet can create an efficient market for public good provision and
•   The ideal conditions:
        •   Zero cost of moving from one community to another.
        •   Zero cost of creating new and duplicate communities.
        •   No spillovers.
•   If these conditions hold then:
        •   Communities will arise to cater to all kinds of public good demands.
        •   People will move to the communities that best satisfy their demand for public goods.
            Creating an ideal match between public good demand and supply.
        •   Production will be efficient because corrupt and/or inefficient public good providers will
            lose customers.
•   In short, under these conditions Tiebout competition duplicates all of the
    best aspects of competitive markets.
                  Tiebout Competition (cont)
•   Aside: The ideal conditions for Tiebout competition are best approximated in
    the real world by virtual communities!
         • Some virtual communities have hundreds of thousands of members who spend
           significant amounts of time living in artificial worlds.
         • Buying and selling of virtual items (e.g. magic swords) in the real world is common.
         • The real word GDP of some virtual communities exceeds that of some small countries.
         • Cyberspace law raises many questions about free speech, the right exclude, the right to
           terminate, ownership of online identities and so forth. Understanding Tiebout
           competition can help to answer some of these questions.
•   What about the real, real world?
•   Consider the rise of private governments. Private governments such as
    homeowners associations and condominium cooperatives provide all
    manner of public goods, from road maintenance, trash collection, and snow
    removal to transportation, policing, and medical care.
         • Practically unheard of in 1960, today some 54.6 million people in the United States live
           in various neighborhood associations.
         • A majority of new housing units in rapidly growing urban areas are privately governed.
           (Local private governments are also becoming common in Britain, Spain, Brazil, and
           even in several former and current communist nations, including Russia and China.)
•   Homeowner associations produce different combinations of public goods
    (pools, transportation, security, medical care etc.) to match the demands of
    their prospective residents.
         • Leisure World communities, for example, cater to retired people with golf courses,
           group recreation activities, and easy access to high-quality health care.
•   Even at the level of county, city and town voting with one’s feet can be
    important. E.g. rise of the new south due to more open labor law.
            Tiebout Competition and Education
 •     Some metropolitan areas include many school districts
       others only a few.
             •    In Boston there are 70 school districts, most within easy
                  commuting distance of downtown, Miami has only 1 school
             •    Choice among districts (which control school revenue and
                  policies) is thus much larger in Boston than in Miami.

•What difference does increased choice make to schooling outcomes?
•Caroline Hoxby (2000) compares school achievement, spending, and productivity
(achievement/spending) with the amount of Tiebout choice across metropolitan regions.
•Hoxby finds that increasing the number of districts from 1 to approximately 10 or so equally
sized districts would:
     •Increase test scores by one-quarter to one-half of a standard deviation – this is a big effect.
           •Most estimates suggest that cutting class size by one-third to one half would not increase test scores by this
           •Hoxby also finds that this increase in choice raises earnings later on in life (presumably through the
           achievement effect) by 15 percent a very big effect.
     •Decrease school spending by approximately 7-8 percent.
           •The decrease in spending with an increase in achievement suggests that in metropolitan areas with little
           choice schools are spending money on things that have little benefit to students (staff, wages, amenities etc.)
     •Increase productivity – since achievement is increasing and spending is decreasing, choice
     increases the productivity of schools.
                Equalization vs Competition
•   Hoxby’s analysis has two important implications.
     – First, if Tiebout competition works then school choice with vouchers would
       almost certainly be even better as vouchers would greatly reduce the cost and
       increase the range of choice.
     – Second, equalization plans can have negative effects.

•   The problem with equalization plans.
•   Competition disciplines school administrators through district spending –
    when choice is possible people move to districts that have good schools
    and this increases the local property tax base allowing districts to increase
    spending (on themselves as well as students!). Similarly, poor performance
    results in a lower tax base and less revenues.
•   Equalization plans switch funding from local property taxes to state income
    taxes (or other state taxes).
•   But Tiebout competition is unlikely to work when revenues are divorced
    from performance.
•   Equalization may equalize everyone at a lower level than competition!
•   Equalization plans may represent a form of collusion designed not simply to
    equalize but to cartelize.

•   Equalization shows up in many areas other than schools. Later we will
    discuss equalization in spending among states in the “federalist” system of
    Argentina (Canada is a similar example).
• Aside: The number of school districts in the United States has declined
dramatically since the middle of the last century.
• Most of the decline is due to an increasingly urban society, especially a
decline in the number of farmers. Nevertheless, the decline in effective
competition may have had negative effects.
        Tiebout Competition=Choice of Law
•   Tiebout competition applies not just to physical space but to “legal space.”
    It’s not necessary to move to the law if the law can move to you.
•   The U.S. has a choice of law approach to corporate charters, which govern
    relations between managers and shareholders. Firms can choose any of
    the 50 states and DC to incorporate, each with its own legal regime.
•   Roberta Romano (1993) argues that this competitive federalist system is
    responsible for the “genius of American corporate law.”
•   What about a race to the bottom? What happens when a state
    reincorporates in Delaware, the leading issuer of corporate charters? If
    there were a race to the bottom share prices would decline but on average,
    share prices increase.
•   Why don’t we get a race to the bottom?
•   Law is a product. When someone buys a house they are buying not just the
    house but also access to the local schools, parks etc. Analagously, when
    investors buy shares they are buying not just the firm’s physical assets but
    also the law that governs that firm.
•   Homebuyers will not pay as much for a home tied to a bad neighborhood.
    Investors will not pay as much for a firm tied to bad law.
•   Similarly, lenders adjust their rates for risk and one such risk is
    management’s lack of accountability generated by bad law.
•   Thus the cost of capital will rise to firms that incorporate in states with bad
•   State competition is not perfect.
         • Concentrated interest groups like managers can bend the
           law to their advantage but competition provides a limit.
           When Pennyslvania changed its regime to strengthen anti-
           takeover provisions many firms opted out, not because the
           managers didn’t like the provisions, but because their
           investors threatened to sell.
         • Delaware is among the least restrictive of hostile takeovers.
         • A monopoly regime reduces the options for exit and thus
           makes it harder for shareholders to discipline firms.
      Federalism, Innovation and Yardstick
“It is one of the happy incidents of the federal system that a single
     courageous State may, if its citizens choose, serve as a laboratory;
     and try novel social and economic experiments without risk to the rest
     of country.”
            Justice Brandeis, New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (dissenting

•   We see competive federalism and innovation at work in corporate
•   Corporate law is continually undergoing changes in response to new
    developments. The states experiment with changes and the best
    changes then diffuse throughout the U.S.
•   Corporate law in the United States advances at a far faster rate than
    in more centralized systems such as in Europe where many of the
    central directives are leftovers from the 19th century long-abandoned
    in the United States.
       Federalism, Innovation and Yardstick
                Competition (cont)
•   Even if mobility is not possible, the existence of other governments
    provides a yardstick to measure the quality of your government.
         • The East Germans knew what life was like in the West and
           this knowledge helped to ferment the revolutions of 1989-
         • Governors who raise taxes are more likely to be reelected if
           neighboring states also raise taxes.
         • Improved policing strategies in New York were adopted by
           other cities when crime fell in New York.
                        Minimal Federalism
•       A minimal definition of federalism (Riker 1964):
          1.   At least two hierarchical governments over the same land and
          2.   Each government with institutionalized autonomy in its own
               •   Institutionalized autonomy means that the autonomy of the
                   subnationals is protected by more than parchment. E.g.
                    •    An upper chamber based on the subnational units rather than
                         population. (Senators were originally elected by the state
                         legislatures. 17th Amendment introduced direct eletion of
                    •    Requiring that constitutional change be endorsed by a super-
                         majority of subnational units.
                    •    Independent courts.
•       Minimal federalism is not sufficient to generate
        important benefits.
    –     Tiebout competition (and yardstick competition) doesn’t do
          much if the field of competition is unimportant.
             Market Preserving Federalism
•   Market preserving federalism (Weingast 1995) is defined as minimal
    federalism plus:
        •   Subnational governments have primary regulatory responsibility over the
        •   Subnational governments cannot erect trade barriers against one another –
            national free-trade is ensured.
        •   Subnational governments face hard budget constraints.
•   MPF means that Tiebout/yardstick competition will be over
    something important, economic policies set by subnational
•   Competition encourages the subnational governments to adopt
    good policies. In addition, bad policies are limited by hard budget
•   Unlike the central government, subnational governments (e.g. state
    governments in the U.S.) cannot print their own money. Thus
    subnational governments must tax their citiziens directly to fund
        •   Since the state governments can’t print money their borrowing rights are
            also typically limited to capital improvements de-facto if not de-jure. Too
            much borrowing for consumption binges and capital costs will increase.
           Market Preserving Federalism (cont)
•   Political pressures will always
    encourage governments to, for
    example, bail-out inefficient
    firms. MPF helps to depolitize
      – The governments with the
         right to bail out inefficient
         firms are limited by
         competition and hard
         budget constraints.
      – The central government
         does not face competition
         and has a soft budget
         constraint but it does not
         have the right to bail out
•   MPF requires real autonomy of
    the subnationals. If the
    subnationals are greatly
    subsidized they are insulated
    from competition.
•   The commitment not to bail out
    the subnationals may not be
    credible when nationals are
    heavily subsidizing
       Market Preserving Federalism (con’t)
•   MPF clearly is inspired by the original Federalism of the United
           The powers delegated by the proposed Constitution to the Federal
           Government are few and defined. Those which are to remain in the
           State Governments are numerous and indefinite. The former will be
           exercised principally on external objects, as war, peace, negotiation
           and foreign commerce;... the powers reserved to the several States
           will extend to all the objects, which, in the ordinary course of affairs,
           concern the lives, liberties and prosperities of the people.
                                            The Federalist, #45, emphasis added.

•   Weingast, notes, that other countries have had market preserving
    federalism if not in law then in practice.
•   A key question (not to be answered here!) is when is MPF stable?
         • Nationals and sub-nationals must be able to check one another.
           Subnationals must have real sovereignty.
         • Parties may be an important stabilizing force if national parties must build on
           local bases.
         • Succesful constitutions limit how much is at stake in politics.
                 Market-Destroying Federalism!
                 Argentina as an example of what not to do.
•   On paper, Argentina has a very decentralized government. The provinces
    have the primary responsibility for education, health services, poverty
    programs, housing, economic infrastructure and so forth and they also have
    the primary rights to collect taxes.
•   Thus, on paper each of the 23 provinces in Argentina is “founded on its own
    bottom” and Argentina could be an example of MPF in action.
•   But in practice, Argentina’s fiscal structure is bizarrely inverted. Spending is
    decentralized but the provinces have ceded taxing authority so that taxes
    are centralized. Taxes flow to a single large bucket in the center before
    being redistributed back to the provinces.
     – About half of the total spending is done by the provinces (two-thirds
        excluding pensions) but a majority of the spending is financed by
        transfers from the center.
     – In fact, most of the provinces finance less than 20 percent of their own
•   What incentives does this create?

                           The parable of the lunch bill.
                                 But it gets worse!
•   Each province has its own central bank and the provinces borrow extensively in both
    the national and international markets. High debt is the result!

•   The Federal government has repeatedly bailed out the provincials. This is both a
    sign of the problem (undisciplined provincial expenditures) and a cause of the
    problem (moral hazard).

•   In short, the provinces do not have hard budget constraints – violating one of the
    principle requirements of MPF.

•   Furthermore, at the same time as the provincials are spending someone else’s
    money they are taxing their own citizens for the benefit of other people! Thus
    expenditures are high but provincial taxes are low.
     –   Why tax the people who elect you in order to benefit the people in another province?

•   At the national level some taxes are required by law to be shared with the provinces
    (and vice-versa) but tax revenue from other sources is non-shared. This leads to an
    inefficient focus on the non-shared taxes which are too high at the same time as the
    shared taxes are too low.
•   Taxes are shared based on an archaic coefficient system and not on the basis of

•   Bottom line is that minimal federalism is not enough. MPF requires the discipline of
    competition - the subnational governments responsible for economic policy must bear
    the costs and benefits of their political choices both economically and politically.
Problems with Tiebout Competition
•   Spillovers/Redistribution
•   Who is on the margin of mobility?
•   The virtues of voice over exit
• When matching is imperfect we say there are spillovers
  (also called externalities). Spillovers may be positive or
       • A locality that dumps sewage into a river imposes a negative
         spillover on downriver communities.
       • A state that supports university research that produces ideas that
         benefit people in other states or even other countries is creating a
         positive spillover.
• Without intervention the community producing the good
  is unlikely to take into account the costs or benefits
  imposed on other communities. It follows that:
   1. Negative spillovers mean too much of the public good is
   2. Positive spillovers mean too little of the public good is created.
• How can spillovers be addressed if matching is too
    Competition, Spillovers and the Difficulty of
•    With spillovers production of public goods will not be optimal but
     worse yet competition can make spillovers worse. For example:

•    Corruption spillovers.
          • Two monopolies in “competition” are more than twice as bad as one. The
            Rhine river problem. Better a king than many grasping princes.
•    Competing for new investment.
          • But note that such competition can be good or bad depending on whether
            the “subsidies” transfer resources or improve efficiency.
•    Gambling?
•    Redistribution and the problem (?) of welfare magnets.
          • British 19th century poor laws tried to limit mobility by restricting the poor to
            their home parishes.
          • Articles of Confederation excluded “paupers” from right to free travel across
            the states.
          • 1996 welfare legislation said benefits to newly-arrived residents could be
            limited to the amount they would have received in the state of exit. But this
            was ruled unconstitutional in the 1999 Supreme Court case Saenz v. Roe.
          • Recall also that welfare reform ocurred only because Presidents Reagan,
            Bush and Clinton encouraged states to apply for waivers to experiment with
            different welfare rules.
                 Wiping up the Spillovers:
               Intergovernmental Relations
•   Coordination: If interests are common sometimes coordination is
    enough to solve problems.
         • Coordination of tax systems.
         • Product standardization.
         • Regulation of air travel.
•   When bargaining is not enough Federalization, i.e. relinquishment of
    control to the central government, may be justified.
         • The federal government “internalizes” the spillovers and so doesn’t have an
           incentive to underinvest or overinvest as would the sub-nationals but lack of
           incentive to do wrong is not the same as a positive incentive to do right!
•   Central government control need not be complete.
         • Intergovernment grants.
         • Matching grants versus block grants.
         • Welfare reform of 1996 moved from matching grants to block grants – block
           grants give the states greater room and incentive for innovation (perhaps
           leading to diffusion via yardstick competion) at price of reduced cost-sharing
           and thus a higher price on the margin for increasing welfare.
         • Note also, however, that the “decentralization” to the states was also
           accompanied by strings.
         • There is a very real danger that central finance becomes central control.
      Who is on the Margin of Mobility?
• Not all citizens are equally mobile.
• Tiebout competition assumes that the mobile
  citizens are similar to the immobile citizens so
  that Tiebout competition will benefit everyone.
      • In the same way that careful shopping by consumers who are
        informed benefits all computer buyers.
• But what if the mobile citizens differ
  systematically from the immobile citizens?
• Tiebout competition will then benefit the mobile
  with no necessary gain to the immobile and
  perhaps at the expense of the immobile.
      • The wealthy elderly.
      • Mobile corporations (subsidy seekers).
                 Is Voice Superior To Exit?
•   If exit is costly then citizens may exert more “voice.” Opening the
    exits can diminish voice to the detriment of third parties. E.g.
         • School choice leading to exit of the types of parents who monitor schools.
         • The rise of ghettos as the black middle class exited the inner cities taking
           their social mores with them – a negative effect of declining racial animus.
         • “Secession of the succesful” and the rise of “gated communities.”

•   But voice and exit can be complements as well as substitutes.
         • The threat of exit makes voice more effective.
         • Voice may be heard across states as when Voltaire fleed to Britain to avoid
           persecution in France.

•   Voice is not always good – it depends on who is speaking!
         • Owners of immobile assets have a greater incentive to dominate the political
           process, precisely because they cannot exit! Agricultural interests, for
           example, are very immobile yet dominate politics in many countries and
   Federalism Around the World
               A Brief Detour

• The US is one of the most decentralized
  (federalist) countries in the world but has
  been slowly centralizing over the past

• Around the world most countries are
  becoming less centralized.
Spending Shares by Federal, State and
Local Government in the United States,
         Countries Around the World are Becoming
                    More Decentralized
• On a 1 to 4 index of political
decentralization the world
average nearly doubled from
1975 to 1995, rising fom 1.03 in
1975 to 1.94 in 1995.
• Developed countries tend to be
more decentralized.
• Latin American countries
decentralized tremendously
since 1980 reaching developed
country levels by 1995.
• Ex Soviet Block countries are
quite highly decentralized.
• The Middle East and N. Africa
are the least decentralized.
• Democratization and
decentralization are highly
         Central Government Share of Spending
                   is Slowly Declining
• The central government
share of spending also
shows trends towards more
• Note, however, that
political measures of
decentralization show more
decentralization than the
central government share of
spending, i.e. more
decentralization on paper
than in practice.
• Latin America is
considerably less
decentralized than the
developed countries on this
                    Federalism and Crime
                         Applying the Tools
•   Prior to the twentieth century the states defined and prosecuted
    nearly all criminal conduct – with exceptions for treason, bribery of
    federal officials, theft of government property and a few other clearly
    federal issues.
•   Theory suggests that this was efficient.
     – Crime is mostly a state and especially a local matter.
     – Preferences and circumstances vary across the United States so that
       the optimal punishment may differ in different regions.
     – Innovations in crime fighting may arise in the laboratories of democracy
       and diffuse throughout the United States (e.g. broken windows.)
     – Consumers care about crime and move from places of high crime to
       places of low crime thus Tiebout competition can beneficially create a
       competitive marketplace in crime fighting.
•   The expansion of interstate commerce and with it interstate crime
    certainly suggests, however, that increasing federalization of crime
    may be justified.
•   But how much federalization is necessary and of what crimes?
Federalism Theory and the Federalization of
•    A federal role in crime control can be justified
     in two circumstances.

    1. Positive Spillover: The states underinvest in crime-
       fighting because the costs of fighting crime are
       theirs alone but some of the gains spillover to other

    2. Negative Spillover: The states overinvest in crime
       fighting because the benefits of fighting crime are
       theirs alone but some of the costs spillover to other
    Federalism, Crime Control and Positive
•   Positive Spillovers
•   Examples:

        A. Who should/will prosecute a hacker who lives in Georgia but who hacks
           into computers in New York City?
        B. Who should prosecute a kidnapper who crosses state lines. (Lindbergh
        C. Crime fighting in border cities to detect and capture imports of contraband
           and exports of stolen property.

    •   How should these spillovers be handled?

    •   Federalization of crime may be best way to handle examples A and B.
    •   In example C, however, federalization is not necessary. Subsidies to
        border cities will also internalize the externality and are likely more
  Federalism, Crime Control and Negative
• Negative Spillovers
• Examples:
     A. Moving the prostitutes out.
     B. Higher penalties (especially for crimes where the criminals
        are mobile, e.g. meth labs, auto theft).
     C. Banishment

  • How should these spillovers be handled?
  • Case A is mostly a spillover of one local area to
    another local area and hence makes a case for state
    internalization and/or local bargaining not federal
         • The spillover should be handled by the smallest government
           capable of internalizing the spillover.
    Federalism, Crime Control and Negative
               Spillovers (con’t)
•   Case B - higher penalties (especially for crimes where the criminals are
    mobile, e.g. meth labs, auto theft) - has the surprising implication that the
    Federal government should tax states with high penalties or force the states
    not to exceed maximum penalties.
     – Recall that if pollution spills over from state A to state B then we are likely to
       have too much pollution. The Federal government should make it more costly for
       the states to pollute – i.e. give the states an incentive to act as if they bore all the
       costs of their pollution.

     – Pushing criminals out-of-state is like pushing pollution out-of-state.

     – The more mobile criminals are the more severe the problem. High penalties on
       wife abuse will not cause many men to move their families to more lenient states.
       But draconian penalties on meth labs may simply move the meth lab from one
       state to another with no benefit to the nation as a whole.
          •   After a state trooper was killed in a meth related event Oklahoma required photo IDs
              and registration to buy pseudoephedrine (an input to meth found in many cold
              medicines like Sudafed). Meth labs declined in Oklahoma but increased in neighboring
              states. A spokesperson for the Kansas Bureau of Investigation said “if all the other
              states do pass this, and you don’t, you’re going to become a magnet for meth cookers.”
          •   It may be efficient to have laws like this in all states but it could also be that once other
              states pass their own laws the gains to Oklahoma do not exceed the costs. It doesn’t
              make sense for Oklahoma to pass these laws if they work only when other states do
              not have them.
    Federalism, Crime Control and Negative
               Spillovers (con’t)
•   Case C: Banishment
•   Surprisingly (to at least this non-lawyer!) banishment is not illegal
    especially as a condition of probation or parole. Banishment may
    sometimes even pass muster as punishment.
         • The Georgia constitution prohibits banishment from the state but in response
           some judges have imposed “158 county banishment.” Banishment from 158
           out of GA’s 159 counties!
•   Banishment from a state is a very clear example of a negative
    spillover which leads to overproduction. Thus, banishment from a
    state should be illegal as a matter of Federal policy.
•   What about banishment from a county?
         • Banishment could serve a useful purpose in removing an offender from
           negative peers.
         • State government can internalize externalities of within-state banishment but
           the GA example indicates that even the power of within-state banishment
           can be abused.
  Federalism, Crime Control and Negative
             Spillovers (con’t)
• Negative crime spillovers are in most cases
  probably not a big deal because criminals are
  not that mobile. Thus, we need not worry too
  much if the Federal government doesn’t try to
  limit state criminal penalties.
• But we should be worried when the Federal
  government increases penalties for crime, i.e.
  the opposite of what theory suggests is optimal.
• In almost all cases the Federal penalty exceeds
  the State penalty for the same or similar crime.
  Federalization of Crime Has Grown Even
           Where Not Necessary
• There are now at least 3,500 federal crimes and
  probably well over 4,000.
• A large fraction, about half, of these have been added
  since 1970.
• Many federal crimes do deal with positive spillovers
  and/or federal questions.
• But many federal crimes followed the wake of news
  flurries about especially brutal crimes and seem to have
  been passed so that Federal officials could be seen to
  be “doing something,” rather than because the states
  lacked any competence or incentive to fully prosecute.
      • Speaking of a series of laws expanding the jurisdiction of the federal
        courts, Justice Rehnquist noted gently “one senses from the context
        in which they were enacted that the question of whether the states
        were doing an adequate job…was never seriously asked.”
        The Crime Fighting Congress
• In 1998 House Majority Leader Dick Armey argued:
   – ‘‘[The] Republican-led crackdown on violent criminals
     . . . is the real reason for recent gains in public
• To support his claim, he cited a long list of ‘‘legislative
  victories,’’ including
   – The Juvenile Crime Control Act
   – The Church Arson Prevention Act
   – The Sex Crimes against Children Prevention Act
• Armey’s argument might leave the naïve to wonder why
  juvenile crime, church arson and sex crimes against
  children were not illegal until 1998.
      Federalism and Medical Marijuana
• In November of 1996 California voters passed the
  Compassionate Use Act which changed the CA code so
  that the possession and cultivation of marijuana for the
  “personal medical purposes of the patient” and “upon
  the written or oral recommendation or approval of a
  physician” was not illegal.
• Ten other states have passed similar laws.
• The Federal government, however, continues to
  prosecute marijuana users under Federal criminal law.
• Yet the theory of federalism is strongly in favor of
  decentralization on this question. Laboratories of
  democracy, Tiebout competition, yardstick competition,
  hetereogeneity of preferences, few spillovers – all these
  suggest that on issues like medical marijuana the states
  should have full authority.
   Federalism and Doctor-Assisted Suicide
• Oregon voters approved Measure 16 the Death With
  Dignity Act in 1994. The act was challenged but was in
  essence passed again with strong support (60-40) and
  took effect in October of 1997.
• The act allows terminally ill patients, under proper
  safeguards, to obtain a physician's prescription to end
      • Over 500 people a year commit suicide in Oregon but only about 30
        of these are terminally-ill patients who ask for a doctor’s
• Federalist principles support the right of Oregon voters to
  pass a doctor-assisted suicide law. Laboratories of
  democracy, heterogeneity of preferences, few spillovers
  – all these suggest that on issues like doctor-assisted
  suicide the states should have full authority.
            Other Problems With the
            Federalization of Crime
•   Overloading of the Federal Courts
     – Chief Justice William H. Rehnquist warned: "The number of cases
       brought to the federal courts is one of the most serious problems facing
       them today....Over the last decade, Congress has contributed
       significantly to the rising case-load by continuing to federalize crimes
       already covered by state laws....The trend to federalize crimes that
       traditionally have been handled in state courts not only is taxing the
       judiciary's resources and affecting its budget needs, but it also threatens
       to change entirely the nature of our federal system."

•   De-facto Double Jeopardy

•   Selective Prosecution. Aristotle said justice means treating equal
    persons equally. What then to make of Rudolph Giuliani’s Federal
•   Federalism has many advantages; matching of public good scope with government
    scope can improve efficiency, Tiebout competition can better match preferences with
    production and can discipline governments, yardstick competition can generate and
    disseminate innovations and so forth.

• But Federalism has no natural constituency!

•   National government wants more power as do the locals. In different times and
    places this can lead to centralization or severe decentralization and sometimes too
    much of both! (As in Argentina, for example).

•   As the examples of medical marijuana, and assisted euthansia indicate even groups
    that ideologically are in favor of “state’s rights” and “decentralization” will quickly
    abandon any principled position once they are in power.

•   Federalism is about a balance of power. Balances are hard to maintain.

          Judges occupy a unique position in a federalist system.
                 Only Judges can maintain the balance!

To top