Bureaucrats or politicians

Bureaucrats or Politicians?∗
Alberto Alesina†and Guido Tabellini‡
First Draft: May 2003; This draft: May 2005
Abstract
Policies are typically chosen by politicians and bureaucrats.
This paper investigates the normative criteria with which to al-locate
policy tasks to elected policymakers (politicians) or non
elected bureaucrats. Politicians are preferable if ability is less
important than effort or there is little uncertainty about whether
the policymakers has the required abilities; if there is uncertainty
about social preferences and flexibility is valuable; if time incon-sistency
is not an issue; if vested interests do not have large stakes
in the policy outcome; if policy complementarities and compen-sation
of losers is important.
JEL classifications: H1 E00 K00.
Keywords: politics, delegation, bureaucracies.
1 Introduction
Policies are chosen and implemented by both elected representatives
(politicians) and non elected bureaucrats. The view that politicians
∗For useful comments we thank an anonymous referee, Philippe Aghion, Timo-thy
Besley, Alessandro Lizzeri, Oliver Hart, Tom Romer, Andrei Shleifer, Charles
Wyplosz and participants in seminars at Harvard, Princeton, Geneva, the CIAR
meeting in Toronto, March 2003, and the Wallis Conference in Rochester in October
2003. This project was initiated while Alesina was visiting IGIER at Bocconi Uni-versity;
he is very grateful for the hospitality. Tabellini thanks CIAR for financial
support.
†Harvard University, NBER, and CEPR.
‡Bocconi University, IGIER, and CEPR.
1

choose policies and bureaucrats implement them is too simplistic; the
boundaries between decision and execution are a grey area and in many
cases bureaucrats do much more than executing either de jure or de
facto. For instance, in most countries non elected central bankers con-duct
monetary policy, with much independence. Regulatory policies are
normally the result of both political and bureaucratic intervention, but
the rise of the regulatory state has made the bureaucracy a key player
in both the decisions and the execution of a large amount of legislation.
Fiscal policy is by and large chosen by elected representatives (govern-ments
and legislatures): bureaucrats are involved in important aspects
of auditing and implementation, but they do not choose tax rates or the
amount of spending for their department. Foreign policy decisions are
made by politicians.
Is this division of tasks appropriate? More generally, what criteria
should guide the allocation of responsibilities amongst politicians and
bureaucrats? We explore this question from a normative perspective by
asking what is the socially optimal allocation of tasks between these two
types of policymakers.
Economists have emphasized one specific argument in favor of del-egation
of policy to a non elected bureaucrat: time inconsistency in
monetary policy. Rogoff (1985) pointed out that an independent and
2

inflation averse central banker not subject to ex post democratic control
would improve social welfare. But there is more to it. For instance, fiscal
policy too is marred with a host of time inconsistency problems, but so-cieties
seem reluctant to allocate this policy prerogative to independent
bureaucrats. 1 An interesting question is why this never happens, and
whether it is justified by normative criteria. An ability to commit to a
course of action may even be desirable in foreign policy, which however
is always the prerogative of appointed politicians, at least in the more
relevant phase of choosing the general strategy.
We focus the analysis on the individuals at the top (party leaders
or high level bureaucrats such as central bank governors). Our premise
is that the main difference between top level politicians and top level
bureaucrats lies in how they are held accountable. Politicians are held
accountable at the elections, for how they have pleased the voters. Top
level bureaucrats are accountable to their professional peers or to the
public at large, for how they have fulfilled the goals of their organization.
1 Blinder (1997) argues that some aspects of fiscal policy could be
allocated to an independent agency operating like an independent Cen-tral
Bank. Also the Business Council of Australia (1999) proposed that
tax policy in Australia be set by an independent agency within limits
imposed by the legislature.
3

These different accountability mechanisms induce different incentives.
Politicians are motivated by the goal of pleasing the voters and hence
winning the elections. Top bureaucrats want to fulfill the goals of their
organization, either because of a "career concern" - they want to appear
competent to improve their external professional prospects in the public
or private sector - or because they draw internal satisfatiction from doing
well whatever they are expected to do.2 Armed with this premise, we
analyze a model of task allocation in which a social planner exploits the
different incentives of bureaucrats and politicians and assigns tasks to
maximize social welfare.
We analyze many different types of policies. From a normative per-spective,
politicians are preferable for tasks that have the following fea-tures:
i) good performance is due to effort more than to ability, and
the required abilities are standard in the sense that there is little uncer-tainty
about the politician’s ability to perform his task. ii) flexibility is
valuable, because social preferences are unstable and uncertain, or be-
2 For a discussion of how bureaucrats are motivated by prospect of
career enhancement and this leads them to internalize the goals of the or-ganization,
see the classic treatment inWilson (1989) especially Chapter
9. In addition, by appearing competent, the bureaucrat can guarantee
his autonomy and independence (Carpenter 2001).
4

cause the policy environment can change rapidly; iii) time inconsistency
is unlikely to be a relevant issue and intertemporal trade-offs are not im-portant;
iv) side payments to compensate the losers are desirable and
relevant, or bundling of different aspects of policy management and a
comprehensive approach is important; iv) the stakes for organized inter-est
groups are small, or law enforcement is weak so that corruption is
widespread.
A recent literature on principal-agent models addresses related issues
in career concerns models. Dewatripont, Jewitt and Tirole (1999a,b)
discuss the foundations of this approach and apply it to study the be-havior
of government agencies. They focus on some issues related to
ours, namely how the nature and ”fuzziness” of the agencies mission
shapes bureaucratic incentives, but they do not contrast bureaucratic
and political accountability. Maskin and Tirole (2001) investigate the
attribution of responsibilities between accountable and non accountable
agents. The latter have intrinsic motivations, while the former seek to
please their principals because of implicit rewards (career concerns). In
our set up, instead, we neglect the role of intrinsic motivations: both
bureaucrats and politicians need to be kept accountable with implicit
incentives; but the implicit incentive schemes can be of two kinds: those
that define a politician (striving for re-election), and those that define a
5

bureaucrat (career concerns). Schultz (2003) contrasts direct democracy,
representative democracy and bureaucratic delegation. Like Maskin and
Tirole (2001), he views bureaucrats as unaccountable and focuses on the
trade-off between ideological polarization and accountability: bureau-crats
are less polarized than partisan politicians, but are more inflexible
since they are unaccountable and cannot be removed after shocks to the
voters’ policy preferences. Besley and Ghatak (2003) also study intrinsi-cally
motivated agents, and focus on how to combine intrinsic motivation
with implicit rewards. Besley and Coate (2003) contrast appointed and
elected regulators of public utilities; both policymakers’ types are in-trinsically
motivated, but direct election allows the voters to unbundle
policy issues. In a related context Hart, Shleifer and Vishny ((1997)
discuss when it is preferable to delegate the provision of public goods
to private enterprises and when to keep it under control of politicians.
Issues regarding incompleteness of the contract between politicians and
private providers have close analogies with some of the questions we
address below.
The paper is organized as follows. Section 2 describes the simplest
case of our model and justifies its assumptions. Sections 3 and 4 discuss
cases of policies with a ”public good” nature and with no redistribution.
Sections 5, 6 and 7 deal with redistribution and with the role of organized
6

interest groups. The last section concludes.
2 The Model
Consider a society that has to decide whether to assign a policy task to an
elected officer or to a bureaucrat. With the generic term ”policymaker”
we indicate who chooses policy, either a politician or a bureaucrat. In the
simplest case we consider a single policy, the result of which is determined
by the effort put in by the policymaker and by his ability. Thus, the
policy outcome y is:
y = θ + a (1)
where a represents the effort of the policymaker and θ ∼ N(¯θ, σ2θ) is his
random ability. Ability and effort are additive.3 Citizens care about the
policy outcome according to a well behaved, concave utility function,
u = U (y). We start with linear preferences, U (y) = y, introducing strict
concavity later when it matters.
Effort is costly, and the strictly convex and increasing cost is labelled
c = C(a). The reward for the policymaker is labelled R(a) and it differs
depending on whether the policymaker is a politician or a bureaucrat.
3 Alternatively they could be multiplicative leading to more compli-cated
algebra but similar results. See Dewatripont, Jewitt and Tirole
(1999b).
7

Both of them maximize their utility defined as:
R(a) − C(a) (2)
with Ca > 0, Caa > 0 and R(a) to be defined below (subscripts denote
partial derivatives).4
The timing of events is as follows. At the ”Constitutional Table” so-ciety
chooses who has control rights over policy, whether the bureaucrat
or the politician. Next, the policymaker chooses effort, a, before know-ing
his ability, θ. Finally, nature chooses θ, outcomes are observed and
the reward is paid. Irrespective of who has control rights over policy,
only the outcome y is observed by the principals, not its composition
between effort and ability. Hence the agent’s reward can only be based
on the policy outcome, y.
In this simple environment, an optimal contract with the policymaker
based on performance would achieve the first best level of effort - see ap-pendix
1. But the assumption that policy performance is verifiable and
contractible is hard to swallow. Public policy typically pursues many
goals, that are often hard to measure and to reward directly through
explicit and verifiable contracts. Moreover, if society could write unre-
4 The model can be restated in terms of rent extraction instead of
effort, by defining a = −r where r > 0 are rents and V (r) (with Vr > 0
Vrr < 0 ) is the utility of rents.
8

stricted optimal performance contracts with its policymakers, then the
question asked in this paper would be utterly uninteresting: bureaucratic
delegation under an optimal contract would always dominate political
delegation. But this implication does not come even close to any ob-served
institutional arrangement.
We thus assume that policy performance, y, is observable but not
contractible. Both bureaucrats and politicians are rewarded based on
observed performance, but through an implicit reward scheme that con-tains
specific restrictions compared to an optimal explicit contract. In
the next two subsections we spell out our specific assumptions about the
implicit rewards offered to a bureaucrat and to a politician, and giving
rise to two different reward functions, RB(a) and RP (a) respectively.
These reward functions are taken as given throughout the analysis. Our
normative question is which reward function is more appropriate, given
the nature of the policy task.
2.1 The bureaucrat
We posit that the bureaucrat is motivated by ”career concerns”. That
is, he is concerned with the perception of his ability θ in the eyes of
those that may offer him alternative job opportunities in the private
or public sector, given the stated goals of the bureaucratic organization.
This assumption is especially appropriate for high level bureaucrats that
9

have already been promoted to the top of the bureaucracy, say a central
bank governor or the chairman of a regulatory agency.5
More precisely, let x be the relevant measure of performance with
which the bureaucrat is evaluated (the stated goals of his organization).
We assume that the bureaucrat’s reward is (the suffix B stands for Bu-reaucrat):
RB(a) = αE(E(θ | x)) (3)
where α is the market value of talent, E denotes unconditional expec-tations
over the random variable x, and E denotes expectations over θ,
conditional on the realization of x. Equation (3) contains several implicit
assumptions. First, the bureaucrat cares about his talent as perceived
by outside observers representing his relevant "labor market". Second,
the expectation of talent is formed by conditioning on the bureaucrat’s
observed performance. Third, the relevant measure of performance, x,
must be defined in advance. Fourth, the market value of talent is a given
parameter, α, possibly different from 1.
In the context of this simple model, it is natural to assume that the
relevant measure of performance for the bureaucrat coincides with social
5 At lower levels of the bureaucracy, job security and promotions dic-tated
by seniority only may imply that maximizing perceived compe-tence
is not particularly relevant for bureaucrats.
10

welfare, so that x ≡ y - this assumption will be relaxed in later sections.
Denoting the public’s perception of a by ae and using (1), we can then
re-write the bureaucrat’s reward function (3) as:
RB(a) = αE(y − ae) = αE(θ + a − ae) (4)
This allows us to easily compute the equilibrium level of effort. First
take the first order condition with respect to actual effort, a, taking
expected effort ae as given. Then, impose the equilibrium requirement
that ae = a. By (4) and (2), we obtain:
α = Ca(aB) (5)
where aB indicates the equilibrium effort of the bureaucrat.
How does equilibrium effort by the bureaucrat differ from that in-duced
by an optimal contract? Comparing (5) with (36a) in section 1 of
the appendix, we see that the bureaucrat puts in the first best level of
effort if α = 1, i.e., if the market value of bureaucratic talent coincides
with the true value of talent for society.6 But if the value of talent for the
bureaucrat differs from that for society, and in particular if it is lower,
then bureaucratic behavior is no longer socially optimal.
6 Here we neglect the bureaucrat’s participation constraint, which
throughout the paper we assume is always satisfied - see section 1 of the
appendix.
11

2.2 The politician
The politicians’s goal is to be reelected and this happens if the voters’s
utility exceeds a threshold W. Denoting by β the value of office, we can
write the reward function for the politician as (the suffix P stands for
Politician):
RP (a) = β Pr(u ≥ W) = β[1 − P (W − a)] (6)
where u = y is voters’ utility and where P (W − a) = Pr(θ ≤ W − a).
Voters are rational. Thus, they realize that the alternative to reelecting
the incumbent is to get another politician with average talent, who will
exert the equilibrium level of effort. It follows that:
W = ¯θ
+ ae (7)
Like the bureaucrat, the politician chooses effort before observing
his talent, taking the voters’ expectations as given. With a normal
distribution for θ, equilibrium effort by the politician, aP , is defined
implicitly by the first order condition:
βn(¯θ
) = Ca(aP ) (8)
where n(¯θ
) = 1/σθ√2π is the density of the normal distribution of θ
evaluated at its mean.7
7 This model could be easily generalized to several periods, if the
12

How does the effort of the politician compare with that of the bu-reaucrat?
Comparing (5) and (8), the answer is ambiguous and depends
on parameters’ values. A higher value of office, β, increases the effort of
the politician, a higher market value for bureaucratic talent, α, increases
the effort of the bureaucrat. Under the assumption that the participa-tion
constraint is always satisfied, in this simple example voters prefer
whatever arrangement results in higher effort. To simplify notation, and
since no additional result hinges on the value of these two parameters,
in the remainder of the paper we set α = β = 1.8
politician’s ability today is a signal of his ability tomorrow but some
random element of ability is present every period so that it can never be
fully learnt in advance. A widely studied case in the political business
ccycle literature is that of a MA (1) process for ability. Persson and
Tabellini (2000) discuss the implications of this political model more
extensively.
A more general formulation, outlined in the appendix, would have the
politician care about both re-election and, conditional on losing office,
his career prospects outside politics. If the value of political office is
sufficiently high compared to the expected benefit of a career outside
politics, then the main implication of our model would still hold.
8 Since we are not considering an optimal contract, both the bureau-crat
and the politician could be earning rents in equilibrium (i.e., their
13

2.3 Discussion
The model seeks to capture a key difference between political and bu-reaucratic
accountability. The politician is held accountable by the vot-ers
who choose whether or not to reelect him, based on their utility. The
bureaucrat is held accountable by his professional peers or by the public
at large, for how he fulfills the goals of his organization. These different
accountability mechanisms induce two behavioral differences between a
bureaucrat and a politician. First, the form of the objective function
differs: the politician strives to achieve a threshold level of utility for the
voters; the bureaucrat wants to maximize his perceived talent. Second,
the relevant measure of performance is different: for the politician it is
the voters’ utility; for the bureaucrat it is whatever goals have been as-signed
to the bureaucratic organization. In this introductory example
only the first difference plays a role, since both voters’ utility and bu-reaucratic
performance are measured by the same variable, y. Hence
the only behavioral difference between the two types of policymaker is
that one maximizes an expected value, the other maximizes a probabil-ity,
both defined over the same random variable. In later sections we
study richer policy environments, where the difference over the relevant
measure of performance also plays an important role.
particpation constraint need not bind).
14

While the assumption that politicians maximize the probability of
victory at the election is now common, there is not a standard model
of bureaucratic behavior. Thus, although we are not the first to use it
(see in particular Dewatripont, Jewitt and Tirole 1999a,b), our "career
concerns" model of a bureaucrat needs some discussion.
Consider first the assumption that the bureaucrat cares about his
talent as perceived by outside observers. While we have justified this
assumption with reference to monetary rewards in future jobs, it can
be interpreted more broadly. Top bureaucrats may care about their
perception of talent "per se", as a matter of self-image, pride or legacy.
Alan Greenspan will probably retire after he resigns from being chairman
of the Fed, but he certainly cares about the perception of his ability in
managing monetary policy.
Next, consider the assumption that the relevant measure of perfor-mance
(from which to infer bureaucratic talent) coincides with the goals
of the organization. If there are multiple tasks, as discussed in the next
sections, then this assumption plays an important role: it does not allow
the bureaucrat or outside observes to select other measures of perfor-mance,
for instance by focusing on tasks where the market value of tal-ent
is higher, or where imperfect monitoring is less of a problem. Thus,
we rule out the case in which, say, a Central Banker chooses to ignore
15

the problem of controlling inflation and, instead, signals his ability in
international relations by publishing speeches and books on that topic.
This assumption can be defended on several grounds. First, as noted
above, a broad interpretation of "career concerns" can incorporate a de-sire
for legacy and good reputation with peer groups, say other Central
Bankers. Second, even taking the "career concerns" literally, future ca-reer
prospects are uncertain and there is a coordination problem: how
does the bureaucrat know which is the relevant measure of performance
used by outside observers? The assumption that performance is assessed
on the basis of the tasks explicitly assigned to the bureaucrat is a nat-ural
focal point to select amongst possible multiple equilibria. Third,
as stressed for instance by Wilson (1989), bureaucratic organizations
have weak internal incentives. To motivate employees, the mission of
the organization must be well defined and pursued by the top bureau-crat.
A leader who is perceived as pursuing his own personal ambition,
rather than fulfilling the organizational goals, is likely to be resisted
by his subordinates and this could undermine the leader’s own perfor-mance.
Moreover, a bureaucrat that pursues his own ambitions rather
than fulfilling the organizational goals would damage his integrity, and
this would certainly hurt his future career prospects. Finally, bureau-crats
(top level or lower ranked) are not chosen at random. Presumably
16

whoever is chosen to perform a specific task has a special ability in that
task, or has a particular motivation to perform it well. This makes it
in his interest to signal ability through that task and not others. In
other words and to continue with the previous example, somebody cho-sen
to be a Central Banker is competent in monetary economics, and it
would not be in his interest to signal his ability in international relations
ignoring monetary policy.
How do these straw men ”politician” and ”bureaucrat” relate to real
world cases? Probably the most compelling example of our ”bureau-crat”
is a Central Banker. His incentives to fulfill his task are mostly
driven by the desire to appear competent, even though even a Central
Banker occasionally may bend to the electoral needs of a ”politician”.
Like our ”bureaucrat”, a Central Banker sets policy without political
interferences and his tasks are set by a clear mandate to keep inflation
low. An American President is instead the quintessential example of a
politician: he seeks reelection for himself in his first term and for his
party in his second, and is not constrained by pre-assigned or narrowly
defined tasks.
Top level bureaucrats in charge of important agencies may be prepar-ing
a leap into politics, so they may worry about their popularity and
not only their competence per se. On the contrary, politicians may look
17

ahead to a career in the private sector. While these caveats point to a
large gray area and intermediate cases between our ”politician” and our
”bureaucrat”, it is useful as a first step to clearly identify how career
concerns and electoral incentives lead to different results depending on
the nature of the policy (but see also footnote 7 above).
3 Imperfect monitoring
We now move to the case of imperfect monitoring, that is a situation in
which performance also depends on extraneous randomness. Thus, we
add noise, ε, besides talent (θ) and effort (a) :
y = θ + ε + a (9)
with ε ∼ N(0, σ2ε
), uncorrelated with θ and unobservable. Only perfor-mance
y is observed and can be the basis of rewards.
In this case the reward for bureaucrats can be rewritten as:
RB(a) = E(E(θ | y)) = ¯θ+ φE(θ + ε + a − ae −¯θ
) (10)
where φ = σ2θ
/(σ2θ
+ σ2ε
) < 1. Given our assumption of normality of the
distributions, we obtain a well known signal extraction result. Now the
perception of talent is ”discounted” by a term φ which reflects the signal
to noise ratio. In equilibrium the choice of the bureaucrat is given by:
φ = Ca(aB) (11)
18

Not surprisingly, the bureaucrat puts in less effort the lower is the signal
to noise ratio.9
Next we turn to political delegation. The politician’s reward is given
by the same expression as above, except that now the distribution from
which the probability Pr(y ≥ W ) can be computed has a larger variance,
that reflects both the variance of θ and of ε. It is immediate to derive
the first order condition of the politician as follows:
n(¯θ, 0) = Ca(aP )
p
σ2θ
where n(¯θ, 0) = 1/(
+ σ2ε
√2π) is the density of the random variable
θ + ε, evaluated at the mean of both θ and ε.
We are now ready to establish the following
9 Note that, with imperfect monitoring, the career concern contract
no longer induces the optimal amount of effort even when there is no
difference between the value of ability for the bureaucrat and for society.
Given risk neutrality, the optimal contract (under the assumption that
the principal only observes y and ability is evaluated equivalently by
society and the bureaucrat) would still induce the same amount of effort
as in (5) above - see also section 1 of the appendix. That is, imperfect
monitoring would not add any distortions. But if the bureaucrat can
only be rewarded implicitly through career concerns, as we assume, then
imperfect monitoring entails an additional loss of welfare for the voters.
19

Proposition 1 The comparison between aP and aB is ambiguous. Im-perfect
monitoring (high σ2ε
) reduces effort for both types of policy-makers.
Higher σ2θ
increases aB but decreases aP .
Therefore, less monitoring does not favor one or the other type of
policymakers. This result is related to those obtained by Dewatripont,
Jewitt and Tirole (1999b), who also point out that performance less
closely tied to talent or effort weakens the incentives of agents motivated
by career concerns. But note that the same conclusions also apply to
a politician. Hence, imperfect monitoring reduces the performance of
both policymaker types (relative to an optimal contract), but it does
not provide an argument for preferring a politician to a bureaucrat at
the constitutional stage.
More uncertainty about talent, however, does favor the bureaucrat
over the politician. With imperfect monitoring a larger variance of
θ increases the effort of the bureaucrat, while it has the opposite ef-fect
on the politician. Intuitively, an increase in the variance of θ in-creases
the signal-to-noise ratio and implies that observed performance
(y) is a better indicator of ability (θ). This makes the bureaucrat work
harder, since by assumption he fully internalizes the benefit of higher
expected ability.10 The politician, instead, only wants to overcome the
10 Here the bureaucrat is risk neutral, which means that his compensa-
20

re-election threshold (giving the voters more than their reservation util-ity
is a waste). If ability is more uncertain (if σ2θ
is high), then re-election
prospects are less sensitive to effort, since more of the policy outcome is
due to randomness. Hence his incentives are weakened.
This result has a practical and sensible implication: bureaucrats are
better than politicians in tasks where the range of possible levels of abil-ity
is wide, that is when there is more uncertainty over the policymaker’s
ability. The reason is not that bureaucrats are more gifted on average,
but rather that they have stronger incentives to pretend that they are
gifted. Very simple tasks are unlikely to be associated with talent un-certainty:
anybody can do them. When tasks become more difficult, the
variance in the level of ability is likely to go up, and bureaucrats are
tion is a linear function of expected ability (conditional on performance).
A risk averse bureaucrat would put in even more effort with more un-certainty
over θ, if his marginal utility was convex (eg. with iso-elastic
utility function, as in the literature on precautionary savings). This
would further increase his attractiveness relative to the politician. But
the opposite would be true if the bureacrat’s marginal utility was con-cave
(in this case more uncertainty over θ could weaken the bureaucrat
incentives, if the effect on marginal utility outweighs the effect on the
signal to noise ratio).
21

preferable to politicians.11
The implication that bureaucratic rather than political accountabil-ity
works better for complex tasks is strengthened if evaluating the per-formance
of a bureaucrat also requires special abilities or skills- that is,
if the extent of imperfect monitoring also depends on who does the mon-itoring.
In the case of politicians, the ultimate judges of performance are
the voters at large. The performance of bureaucrats, instead, is mainly
evaluated by their professional peers. Hence, imperfect monitoring is less
of a problem if politicians are given simple tasks, since bureaucrats can
more easily be held accountable by their peers for more technically de-manding
tasks. Maskin and Tirole (2001) and Epstein and O’ Halloran
(1999) reach a similar conclusion in different models.
Is the real world attribution of task broadly consistent with this im-
11 As pointed out by a referee, bureaucrats also work harder than politi-cians
if performance is more sensitive to ability than to effort. Rewriting
(9) as y = Kθ + a + ε,where K is a parameter that captures the rela-tive
importance of ability, we obtain that a higher K increases aB but
reduces aP . To the extent that ability (rather than effort) is needed in
complex tasks, this reinforces our conclusion. But many complex policy
decisions, such as in foreign policy, require ability of a general rather
than a specialized kind.
22

plication? If difficult tasks are also technically more demanding, then the
answer is clearly positive. In many cases technical tasks are delegated
to bureaucrats: for instance managing the financial structure of public
debt, or regulating public utilities or other industries, while politicians
retain the technically less demanding task of setting general targets. In
the UK, for instance, politicians choose a target level of inflation; the
technically demanding task of choosing interest rates to achieve such
target is delegated to the Central Bank. It is not always true, however,
that difficult tasks are technically more demanding. Some complex pol-icy
decisions, such as in foreign policy, require ability of a general rather
than a specialized kind. According to Proposition 1, these complex and
yet technically not demanding tasks are also better left in the hands of
bureaucrats. But here, we often observe a politician in charge. The next
section suggests a different reason, unrelated to variance in ability, why
politicians may perform better in such policy environments.
4 Policy tasks in an uncertain world
We now add an element of uncertainty to social welfare. In particular,
suppose that at the Constitutional Table voters are not sure of howtheir
preferences will evolve. We return to the case of perfect monitoring and
we assume that there are two possible policies, that is two different
23

directions in which effort can be devoted to: yi = θ+ai, with i = 1, 2.12
With multiple tasks, which will be our focus from now on, one needs
to specify a general cost function with multiple arguments, c = C(a1, a2).
Instead of using the general formulation, we simplify to either an additive
case (c = C(a1+a2)), where effort in the various tasks is perfectly substi-tutable
in the cost function, or to a separable case (c = C(a1)+C(a2)),
where the marginal cost of effort in one task is totally independent of
effort devoted to the other tasks. We choose the simplest formulation
that does not produce knife-hedge or ”trivial” results. The more general
specification of costs generates qualitatively similar results. We begin in
this section by considering additive costs, so that c = C(a1 + a2).
At the Constitutional Table the (identical) voters are uncertain about
their ex post preferences over alternative policies, so that voters utility
is now given by the following concave function:
U (λy1 + (1 − λ)y2) (12)
where λ = 1 with probability q > 1/2, λ = 0 with probability (1 − q).
Thus, society does not know ex ante what it will like ex post; but there
is no disagreement ex post amongst members of society. Disagreements
and redistribution will be analyzed below. The timing is now as follows.
12 For a general discussion of multi task functions in a principal- agent
relationship see Holmstrom and Milgrom (1991).
24

First, at the Constitutional Table voters choose whether to assign this
policy to a bureaucrat or to a politician. Then nature chooses λ, that is
social preferences are determined. Having observed λ, the policymaker
chooses [ai] , then nature chooses θ, and finally policy is determined and
rewards paid. We assume that λ is not verifiable.
Consider bureaucratic delegation first. As discussed in section 2, if
the Constitution assigns control rights over policy to the bureaucrat, it
also defines a relevant measure of performance with which his ability
is evaluated. In that section, social welfare was the natural measure
of performance, because it coincided with the only possible measure of
performance. But here, at the Constitutional Table social preferences
are not yet known, since they depend on the future realization of the
random variable λ. Thus, we assume that the bureaucrat can only be
assigned an unconditional measure of performance, defined as:
x = δy1 + (1 − δ)y2 (13)
where δ is a parameter specified by the Constitution. This formulation
entails two assumptions. First, we assume that the relevant measure of
performance assigned to a bureaucrat cannot be ex-post social welfare,
u. We can justify this assumption with the argument that social welfare
cannot be operationally described ex-ante in an unambiguous way. To do
so, we would need specific assumptions about utility functions, technol-
25

ogy, and many other unforeseable but relevant features of the economic
environment. Of course, individual welfare can be observed ex-post by
polling each individual about the policymaker’s performance. But telling
a bureaucrat that his performance would be assessed ex-post through an
opinion poll would transform him into a politician, and the theoretical
distinction between political and bureaucratic accountability that is at
the core of this paper would be lost. The second crucial assumption is
that the operational and describable measure of performance that can
be assigned to a bureaucrat, and in particular the parameter δ, cannot
be contingent on the realization of the random variable λ : the mission
for the bureaucrat cannot be contingent on the realization of ex post
voters’ preferences. This element of contract incompleteness is plausi-ble,
and again can be justified with reference to the undescribability or
unforseeability of future states of the world.13
Under these assumptions, the rewards of the bureaucrat are:
RB(a) = E(E(θ | x)) = E(θ + δa1 + (1 − δ)a2 − δae
1 − (1 − δ)ae
2) (14)
Given additive costs and q > 1/2, it is optimal for society to set δ = 1.14
13A related structure of contract incompleteness also underlies the models of Con-stitutional
choice by Aghion and Bolton (2003) and Aghion, Alesina, and Trebbi
(2004, 2005). See Maskin (2001) for a general discussion and criticical evaluation of
the assumption of contract incompleteness.
14 If costs were separable, then the optimal δ would be increasing with
q, at a rate that is decreasing with the curvature of U (.) for obvious
26

The first order conditions for the bureaucrat then imply:
aB1
= C−1
a (1), aB2
= 0 (15)
That is, the bureaucrat focuses all his effort on the ”main” activity of
his mandate because that is more helpful in signalling his ability. Thus,
the voters’ utility in equilibrium is given by:
U B = qEU (θ + aB1
) + (1 − q)EU (θ) (16)
The key here is that by choosing a bureaucrat who is non responsive to
the ebb and flows of society’s preferences, citizens are ”stuck” with the
risk that effort is misallocated and the bureaucrat pursues the wrong
goals, those that ex-ante seem more likely to be relevant.
Next, suppose that, at the Constitutional Table, society gave control
over policy to a politician. To win re-election, the incumbent must show
that he is more competent than the opponent, given that voters observe
their own utility. This means giving voters a sufficiently high utility.
Whatever beliefs the voters entertain about effort allocation, and given
that effort is not observed by voters, the politician always finds it in his
own interest to put effort in the task preferred ex-post by the voters.
Thus, if λ = 1, then the politician sets a2 = 0; and viceversa he sets
reason having to do with risk aversion. The qualitative nature of our
result would not change.
27

a1 = 0 if λ = 0. Effort in the chosen task is then determined by a first
order condition similar to (8) above.
This is what differentiates the politician from the bureaucrat. The
politician’s goals always depend on the realization of λ (i.e., on the ex-post
preferences of the voters). The bureaucrat instead must be told
what to do and in some cases he will be assigned the wrong mission.
The following proposition follows.
Proposition 2 The politician, unlike the bureaucrat, always chooses
the right task from the voters’ perspective. This advantage of the
politician is more important the more risk averse are the voters
and the more uncertain are their ex-post preferences.
Delegation to a bureaucrat is safe when society’s preferences are well
known and stable. But when they change, the ”rigidity” of a bureau-crat’s
behavior makes the latter much less attractive. This helps us to
understand why monetary policy is often delegated to an independent
central bank, while foreign policy is typically under the control of politi-cians.
Few would disagree with the statement that the appropriate goal
for monetary policy is to keep inflation under control with some room for
stabilization policy; and this goal is unlikely to change over time. But
preferences regarding foreign policy are unlikely to be stable and un-changed,
and as a result an appropriate simple bureaucratic goal cannot
28

be stated once and for all15.
In these situations, a combination of politicians and bureaucrats
could be welfare improving. In fact, a natural remedy to the ”narrow-mindedness”
of bureaucrats pursuing the wrong task is to let the politi-cian
decide the mission of the bureaucrat. Specifically, the constitution
could prescribe that policy be delegated to a bureaucrat, but the bureau-crat’s
mission (the parameter δ in (13) above) be chosen by a politician.
If the politician observes the contingency λ and if he is held account-able
by the voters as described in the previous section, he would always
choose the socially optimal mission for the bureaucrat. This division of
tasks (the politician assigns the bureaucrat some goals and the latter
chooses the instruments with which to pursue them) is observed in a
variety of real world arrangements. Of course, the precision and fre-quency
with which bureaucratic goals are defined can vary from case to
case, and determines the extent to which an independent bureaucrat is
really in charge of policy decisions (rather than taking orders from the
politician).
There is a case in which ex post flexibility is in fact a disadvantage.
When society’s preferences are time inconsistent, the benefit of flexibil-ity
associated with political delegation has a cost. Politicians are much
15Hart, Shleifer and Vishny (1997) and Wilson (1989) make a similar argument to
clarify why it would be close to impossible to privatize foreign policy or to delegate
it fully to a non-political agency.
29

more likely to fall in the trap of time inconsistency, compared to bu-reaucrats.
The reason is that the goals of a politician are unavoidably
linked to the ex-post welfare of voters, through reelection motives. The
rigidity of bureaucratic control, instead, offers protection against time
inconsistency. The bureaucrat can be given an explicit mission, possibly
different from whatever is ex-post optimal for the voters. This possibility
of strategic delegation enables society to overcome credibility problems
and has been used extensively in monetary policy (Rogoff 1985).
A related issue has to do with the time profile of costs and benefits
of policy choices. Bureaucrats tend to care more about the long run
consequences of policies, compared to politicians, for two reasons. First,
often bureaucrats are appointed for longer than electoral cycles, precisely
to avoid short-termist policies. Second, even when bureaucrats have
short terms of office, the blame for myopic policies may reach them and
hurt them later on. The reason is that bureaucrats care about their
professional reputation in the eyes of their peers. This gives bureaucrats
a strong incentive to focus on the long term goal. When the short-termism
of politicians is an issue, the interaction between bureaucrats
and politicians can yield welfare improvements.16
16 The working paper version of this paper discusses more formally the
benefits of delegation in controlling time inconsistency or shor-termism.
A large literature, surveyed in Persson and Tabellini (2000), models
30

5 Compensation of losers
A critical task of politicians is to form coalitions in favor of certain
policies, compensating losers either with direct transfers or by bundling
several policies into one package. To illustrate this point, we need a
conflict of interest between voters (or groups of voters) and the possibility
of side payments and of bundling policies with complementarities.
Voters’ utility now depends on the policy outcome and a transfer
(positive or negative) received by the government. We have two voters
(or homogeneous groups of voters of equal size) with strictly concave
utility defined over private consumption, U (ci), i = 1, 2 where:
c1 = y1 + t, c2 = y2 − t, y2 ≥ t ≥ −y1 (17)
Therefore t is a direct lump sum transfer between voters and the gov-ernment
budget is balanced. Each group benefits from different tasks
requiring specific and uncorrelated abilities, θi, i = 1, 2. Let the distribu-myopic
electoral cycles in monetary and fiscal policy with rational vot-ers.
Besley and Coate (2003) find evidence that, in US states, elected
regulators tend to keep lower electricity prices compared to appointed
regulators. If, as likely, lower prices come at the expenses of lower in-vestments,
this finding is consistent with the prediction of short-termism
by elected (as opposed to appointed) regulators.
31

tion of θi have the same densities n(.) and cumulative distributions N(.)
(not necessarily normal). There are random negative spillovers between
the two tasks, such that:
y1 = θ1 + a1 − λκa2, y2 = θ2 + a2 − (1 − λ)κa1 (18)
The parameter 0 < κ < 1 denotes the strength of the negative spillover
effects. Who is hurt by the spillovers is ex ante uncertain. Thus, λ
is a random variable that can equal 1 or 0 with equal probabilities.
As in section 4, we assume that λ is observable ex-post, but it is not
describable ex-ante, so that the bureaucrat’s mission cannot be defined
contingent on λ. The policymaker maximizes its usual payoffs, with
different rewards for the two types of policymakers, except that now we
assume that the cost function is additive in the two efforts:
R(a1, a2) − C(a1) − C(a2) (19)
Timing has the usual structure. First nature sets λ and this deter-mines
which group is hurt by the spillover effect. Then the policymaker
chooses ai and t, nature sets θi and rewards are paid.
Consider the politician first. He maximizes reelection probabilities,
which means that he has to win the favor of a strict majority of voters.
Here this means winning the votes of both groups (as it will be clear be-low,
nothing of substance hinges on the fact that in this simple example
32

reelection requires pleasing all voters). Therefore:
RP (a1, a2) = Prob(U (c1) 1 W1) ∗ Pr ob(U (c2) 1 W2) (20)
where Wi is the reservation utility of group i.
Suppose for concreteness that λ = 1. If the two reservation utilities
are equal, then the politician sets transfers t so that:
n(z1)
1 − N(z1)
=
n(z2)
1 − N(z2)
(21)
where z1 = U −1(W )−t−a1+κa2 and z2 = U −1(W)+t−a2. That is, the
politician equalizes the ”hazard rates” of losing votes from either group.
In this context, the hazard rate measures the elasticity of the probability
of winning with respect to transfers. Thus, this optimality condition is
similar to the Ramsey rule of optimal taxation: transfers are allocated
between groups so as to equalize this elasticity across groups. If the
hazard rate is monotonically increasing in z, and given the assumption
of the same distribution for θi, i = 1, 2, equation (21) implies c1 = c2.17
That is, the politician implements full insurance, fully compensating the
losers from the negative externality.
Exploiting (21), the optimality conditions for the allocation of effort
17 A uniform distribution of θ satisfies the assumption of a monotoni-cally
increasing hazard rate, for instance.
33

to the two tasks imply:
n(z1)(1 − N(z2))=Ca(aP1
) (22)
n(z2)(1 − N(z1))(1 − κ)=Ca(aP2
)
Thus, the politician allocates effort ”correctly”, in the sense of devoting
more effort to the task that does not have negative spillovers: aP1
> aP2
if λ = 1. Comparing (22) with (8) in section 2, however, we see that the
politician is induced to put less effort in both tasks, including the one
without a negative externality (task 1), relative to the simple case of only
one task. The reason is that bundling of two tasks requiring different
abilities weakens his incentives. His likelihood of reelection now depends
on his success in both tasks. Even if he puts a lot of effort in task 1, he
could still loose the election because he happens to be unable in task 2.
His awareness of this risk (captured by the term (1 − N(z)) < 1 on the
left hand side of (22)), dilutes his incentives.18
Let’s now turn to the bureaucrat. As in section 4, we assume that
the measure of performance that he is assigned at the Constitutional
Table (and on the basis of which is career-incentives are determined)
cannot be contingent on λ and cannot be formulated in terms of social
welfare (U (c1) + U (c2)) because it is too vague a concept, or cannot
18 Persson and Tabellini (2000) and Seabright (1996) make a similar
point in comparing centralized vs decentralized arrangements.
34

be observed by outsiders to infer the bureaucrats’ talent. With this
restriction, the natural measure of performance in this context is total
output, x = (y1 + y2). If given this goal, the bureaucrat allocates effort
efficiently, taking the negative externality into account:
1=Ca(aB1
) (23)
1 − κ=Ca(aB2
)
Comparing (23) with (22), we see that the bureaucrat puts in more
effort than the politician, since his incentives are not diluted by the
risk of losing the election (the terms (1 − N(z)) are missing from (23)).
Nevertheless, compensating transfers are set to zero.
Comparing the politician and the bureaucrat, we thus have:
Proposition 3 The politician provides side payment to compensate losers
but has weaker incentives than the bureaucrat; the latter, however, does
not compensate losers.
This result follows from the assumption that bureaucrats cannot be
given state contingent missions. If their goal is formulated in terms of
aggregate efficiency, they will neglect the distributional consequences of
their actions. A politician instead can take advantage of relatively com-plex
and evolving spillovers between issues and build majorities with
complex side payments schemes. Compensating the losers makes it eas-
35

ier to pass legislation while at the same time providing insurance against
bad luck. Imagine a policy that favors a large majority, say a badly
needed highway, but that creates losers, say the property owners. Under
democratic choice, the losers might be able to block the project. But the
politician can put together a package of compensation for the property
owners, with large benefit for the majority. In a sense this is almost
what describes the job of a politician. Instead, it is hard to imagine how
a bureaucrat might do that. How can one write on paper what a bureau-crat
is allowed to do or not do, to create bundling and compensation? A
bureaucrat can be delegated the task of building the best possible high-way
and he may potentially do a better job than the politician; but he
does not have the ability, interest or authority to provide compensation
to the local owners.
6 Lobbying and bribing
In this section we consider the case of lobbies that can influence the
choice of policies with bribes or campaign contributions. Thus here
”redistribution” is intended as favors towards powerful minorities that
can influence policy decisions. Both the politician and the bureaucrat
can be captured by the interest group, but with different mechanisms.
This difference can give rise to a constitutional preference for one or the
other type of policymaker, depending on the circumstances.
36

As in section 4, there are two tasks, yi = θ +ai, i = 1, 2 and the cost
of effort is non-separable: c = C(a1 + a2). Task 1 benefits the voters at
large, while task 2 only benefits a small but organized interest group.
Voters influence policy only through elections. The organized interest
group is small and its vote is irrelevant; but he can influence policy
through bribes, b, or campaign contributions, f . Thus, the preferences
of voters are just y1, while those of the interest group can be written as:
(1 + γ)y2 − b − f (24)
where γ is a parameter capturing the intensity of the group’s preferences
for task 2.
Bribes can be offered to both the politician and the bureaucrat, but
are illegal. Thus, if a policymaker accepts a bribe, with some exogenous
probability q he is caught and pays a fine Z (the interest group is not
fined). Campaign contributions are legal and can only be offered to the
politician. The effect of campaign contributions is to increase the incum-bent’s
chances of winning the elections. We model this by saying that
the voters’ reservation utility is a decreasing function of the campaign
contributions collected by the incumbent:
W = ¯θ
+ ae
1 − H(f ) (25)
where the function H(.) captures the effect of campaign contributions.
37

It is natural to assume that H(0) = 0, Hf > 0, Hf f < 0. At the Consti-tutional
Table the lobby has no influence, so if the bureaucrat is given
control rights over policy, his assigned measure of performance coincides
with the task that benefits voters at large (x = y1).Under these assump-tions,
we can write the policymaker’s preferences as:
R(y1, y2) − C(a1 + a2) + b − qZ (26)
where R(y1, y2) are the policymaker’s rewards (RB(y1, y2) = E(θ/y1) for
the bureaucrat, RP (y1, y2) = Pr(y1 ≥ W ) for the politician). The pol-icymaker’s
effort devoted to task 2 is observable by the interest group,
so that bribes and campaign contributions can be contingent upon the
policymaker effort: b = B(a2), f = F (a2). The timing of events is as fol-lows.
First the Constitution allocates control rights over policies. Then
the organized group commits to bribes and or campaign contributions,
as a function of effort. Next, the policymaker allocates effort between
the two tasks. Nature then chooses a realization of θ. Finally, rewards
are paid.
This is a common agency game, with two types of principals: the in-terest
group and the representative voter. The interest group has all the
commitment power and can either influence the agent directly (through
bribes), or indirectly (through campaign contributions). The distinc-tion
between the politician and the bureaucrat is that the latter can
38

only be influenced by the interest group through bribes. We want to
know whether the voters are better off with the bureaucrat or with the
politician, and what influences this comparison.
6.1 Bribing the bureaucrat
If the constitution gave all control rights to the bureaucrat we would
have a standard common agency game, with a single active lobby. If
bribes are positive, then the equilibrium must be jointly optimal for the
organized group and the politician. This immediately implies:
aB1
= 0, aB2
= C−1
a (1 + γ) (27)
Moreover, restricting attention to truthful contribution (here brib-ing)
schedules, the equilibrium bribing schedule has the following simple
form:19
B(a2) = ¯B
+ (1 + γ)a2 (28)
where the constant ¯B
is chosen by the organized group so as to leave the
bureaucrat indifferent between accepting or rejecting the bribe. Given
the bureaucrat’s preferences, this implies:
¯B
= C(aB2
) − C(âB1
) + âB1
− (1 + γ)aB2
+ qZ (29)
where âB1
a (1) denotes the equilibriumpolicy if no bribe is accepted.
= C−1
19 See Grossman and Helpman (2001).
39

Finally, the organized group must also prefer to pay the bribe rather
than be passive. This in turn puts an upper bound on the constant ¯B
that the organized interest group is willing to pay. Taking into account
(29), an equilibrium with positive bribes exists only if the following con-dition
is satisfied:
(1 + γ)aB2
−
£
C(aB2
) − C(aB1
) + aB1
¤
≥ qZ (30)
If instead this condition is violated, then bribing does not take place and
the equilibrium with the bureaucrat delivers the optimal policy for the
voters. Equation (30) makes it clear that an equilibrium in which the
bureaucrat is bribed is more likely if the stakes for the organized group
are high (γ is large), or if the legal system works poorly (qZ is small).
6.2 Lobbying the politicians
Next, suppose that the politician is in charge of policy. A condition
similar to (30) above determines the existence of an equilibrium with
bribes (the expression is not identical because the politician’s reward
occurs through reappointment). In particular, it remains true that bribes
would be zero if the legal system is strong, so that the probability of
being caught is high. But now, besides bribes, the organized interest
group can also resort to campaign contributions. He chooses to do so if
campaign contributions are sufficiently effective in swaying the voters.
Specifically, in an equilibrium with campaign contributions, the al-
40

location of effort must be jointly optimal for the politician and the or-ganized
group, given voters’ expectations. In particular, the outcome
must be optimal for the lobby, subject to the constraint of leaving the
politician indifferent between accepting the campaign contribution and
pleasing the lobby, or refusing the campaign contribution and allocating
effort as optimal for the politician, given voters’ expectations. Let ¯W
de-note
the politician’s utility if it refuses the campaign contributions, given
voters’ expectations. Then the equilibrium must solve the following op-timization
problem by choice of a1,a2 and f, subject to non-negativity
constraints on the three choice variables, and taking voters’ expectations
ae
1 as given:
Max { (1 + γ)a2 − −f} s.to : Pr(θ ≥ ¯θ
+ae
1−a1−H(f ))−C( a1+a2) ≥ ¯W
(31)
The appendix describes the full equilibrium. Its properties depend
on how effective are campaign contributions in swaying the voters - i.e.
on the slope of the function H(f ). If Hf (0)(1 + γ) < 1, then the
equilibrium has zero lobbying (f = 0) and the outcome is optimal for
the voters (aP2
= 0). In this case, campaign contributions cannot be
productive enough, and the organized group will not seek to influence
the politician: the group’s stakes are too low relative to how much he
would have to pay into the electoral campaign of the politician.
41

The opposite extreme occurs if Hf (f ∗)(1+γ) > 1, where f ∗ denotes
equilibrium campaign contributions. In this case, campaign contribu-tions
are very effective at the margin. Effort is allocated entirely to
please the organized group only (a1 = 0), and it is determined jointly
with equilibrium campaign contributions, by the requirement that the
politician is indifferent between accepting or not the contributions and
by the following optimality condition:
n(¯θ
− H(f ∗)) · Hf (f ∗)(1 + γ) = Ca(aP2
) (32)
where n(z) is the normal density of θ evaluated at the point z. For this
to be an equilibrium, the organized group must benefit relative to the
option of not lobbying at all, and this also requires: (1 + γ)aP2
≥ f ∗.
For intermediate properties of the slope Hf (.),the equilibrium could
entail positive effort by the politician on both tasks. Note that in this
case too, voters are hurt by lobbying: given our formulation of the cost
function, effort devoted to please the lobby (a2) reduces effort devoted
to please voters (a1).
We summarize this discussion in the following:
Proposition 4 Political lobbying can be an equilibrium, even if bribes
to the bureaucrat are not. This is more likely if campaign contri-butions
are effective in influencing the voters, but the legal system
is strong and effective in discouraging bribes.
42

Thus, politically appointed policymakers are more easily captured by
organized interests compared to bureaucrats, particularly in advanced
democracies with a well functioning legal system. The reason is that,
to influence a bureaucrat, the organized group needs to engage in illegal
activities and fight against possibly deeply entrenched professional goals
and standards of a technical bureaucracy. To influence a politician,
instead, the interest group has an additional instrument: he needs to
convince the voters that the politician is doing a good job and deserves
to be reelected. The politician will then automatically respond with
policy favors to the interest group, since this will help his chances of
reelection. Thus, policies where the stakes for organized interests are
very high, or where redistributive conflicts concern small but powerful
vested interests against the voters at large, are more safely left in the
hands of the bureaucrat. The regulation of public utilities is a typical
example: the interests of consumers are easy to identify and protect
through regulation, while the stakes for the utilities’ supplier are very
high and a politician may be easily captured.20
Note that this result points to an important difference between ad-
20 This normative argument in favor of bureaucrats is mitigated if they
are easier to bribe than the politician, however. And bureaucrats with
technical expertise may be more easily bribed than politicians through
a "revolving door policy" - i.e. at the end of their public services poli-
43

vanced and less advanced societies. In advanced societies with a well
functioning judicial system, it is relatively easy to enforce the no bribe
equilibrium, but campaign contributions may still be very effective at
buying policies; hence, bureaucratic delegation works well. In develop-ing
countries, instead, stopping bribes might be close to impossible and
politicians are likely to do as good a job as bureaucrats.21
7 Splitting the cake
We now consider a purely redistributive policy, ”cake splitting”. Con-sider
three voters, the minimum number required to make the problem
interesting. The policy task delivers a ”cake” that can be divided be-tween
the three voters, therefore:
y = θ + a = c1 + c2 + c3 (33)
We start with risk neutral voters, that have utility function U (cJ) = cJ ,
and comment below on how the results would change if they are risk
averse.
The key difference between a politician and a bureaucrat is, once
again, that the former needs a majority to win and the latter simply
cymakers are offered lucrative jobs in the private sector.
21 Glaser and Shleifer (2003) reach a similar conclusion, using a differ-ent
analytical framework.
44

wants to signal talent. Consider the bureaucrat first. At the constitu-tional
stage, the bureaucrat can either be given no redistributive tasks,
in which case redistribution is entirely arbitrary - we call this an "unfair"
bureaucrat. Alternatively, behind a veil of ignorance he can be assigned
the task of redistributing equally, that is y/3 for all three voters - we
refer to this case as a "fair" bureaucrat. But irrespective of whether he
is ”fair” or "unfair" (i.e., of how he splits the cake), his talent is still
judged by the aggregate measure of performance, x ≡ y, not by how he
redistributes. His first order conditions are thus identical to those in (5),
section 2.
Next, consider the politician. Since he only needs to please a ma-jority,
he gives y/2 to two voters and zero to the third one. Hence, his
reward is:
RP (a) = Prob(y/2 ≥ W ) (34)
where W is the reservation utility of individual voters. Implicit in (34) is
the assumption that voters expect that the incumbent, if re-elected, will
maintain the same redistribution observed today - i.e. he will split the
cake in half between the voters who re-elect him. With forward looking
and rational voters, W equals the average expected utility they can get if
the opponent is elected. If the hypothetical redistribution implemented
by the opponent is unknown, then W = (θ + ae)/3. Going through the
45

usual steps, of maximizing with respect to effort for given expectations
and then imposing rational expectations, in equilibrium the politician’s
optimality condition implies:
n
μ
2θ − aP
3
¶
= Ca(aP ) (35)
where n(z) denotes the normal density evaluated at point z. Comparing
(35) with (8) in section 2, we see that once the politician is also in
charge of redistribution, he can get away with less equilibrium effort.
The reason is that here he only needs to please two voters out of three.
He can thus reduce effort, and still please two voters with the portion of
the cake taken away from the minority.22
Note the asymmetry: voters expect the incumbent to preserve the
observed redistribution over time, but they are uncertain about how the
opponent would redistribute. This asymmetry creates an incumbency
advantage and dilutes the politician’s incentives: the voters are more
willing to reappoint the incumbent even if he is incompetent, because
they benefit from his redistribution.23 Here we assumed a very stark
asymmetry: no uncertainty at all about how the incumbent will redis-
22 This result is similar to that obtained in Ferejohn (1986) and Persson
and Tabellini (2000). But since here voters are forward looking, we rule
out the Bertrand competition among voters that instead features in the
backward looking voting equilibrium of Ferejohn (1986).
23 Indeed, if the voters were certain to be included in the winning
46

tribute, and maximal uncertainty about the opponent. But the nature
of the results would be preserved with less stark assumptions, as long as
voters are more uncertain about the redistributive policies of the oppo-nent
compared to those of the incumbent.
The assumption that the opponent’s future redistributive policies are
more uncertain than those of the incumbent can be derived from more
primitive assumptions. For instance, suppose that politicians have lex-icographic
preferences: first they care about re-election, as spelled out
above. Second, conditional on being re-elected, they also care about the
welfare of specific groups of voters. Suppose further that voters ignore
these redistributive preferences. Then, the incumbent’s redistributive
policies reveal his preferences, and voters correctly expect these policies
to be continued if he is re-elected. As they cannot observe what the op-ponent
would do, voters face more uncertainty if voting for the opponent.
This simple example also points to the fact that it is in the interest of
politicians to pretend that they are ideologially biased in favor of specific
groups or policies, even if in reality they are purely opportunistic. The
ideology of politicians is like their brand name: it keeps voters attached
coalition by the opponent, their reservation utility would be W = (θ +
ae)/2. In this case the effort of the incumbent would coincide with (8)
and there would be no dilution of effort due to redistribution.
47

to parties and reduces uncertainty about how politicians would act once
in office.24
Given these results, who is better for the voters behind the consti-tutional
veil of ignorance, the bureaucrat or the politician? If voters
are risk neutral, and given that they ignore the redistribution chosen
by the politician, they only care about aggregate performance, y. This
makes the bureaucrat more attractive for the voters for a larger range
of parameter values, compared to the case of simple non-redistributive
tasks in section 2. With risk averse voters, the normative comparison
between bureaucrat and politician also depends on whether the bureau-crat
is "fair" or "unfair". A "fair" bureaucrat is even more attractive
compared to the politician, not only because he is likely to put more
effort, but also because he is less risky - the politician exposes the voters
to the risk of being in the minority.25 But the result may be reversed if
the bureaucrat is "unfair" and implements a totally arbitrary redistrib-ution.
In this case, political redistribution is less risky, since two voters
out of three are always included in the winning majority. The case of
24Drazen and Eslava (2004) analyse a model of electoral policy cycles where voters
infer the redistributive preferences of the incumbent from the policies he enacts.
25 Maskin and Tirole (2001) also point out that the ”tyranny of the
majority” or the expropriation of minorities is one reason why politicians
may do worse than non-elected officials (unaccountable ”judges” in their
context).
48

an "unfair" bureaucrat seems more plausible, since in a complex world
it is difficult to precisely assign redistributive task to a bureaucrat.
We can summarize this discussion in the following:
Proposition 5 The possibility of redistribution reduces the equilibrium
effort of the politician, but not that of the bureaucrat. Risk aversion
makes the bureaucrat more or less desirable ex-ante depending on how
easy it is to impose fair treatment of all voters in his task description.
8 Conclusions
Our analysis rests on two fundamental assumptions. The first one con-cerns
the motivation of different types of policymakers. Bureaucrats
want to signal their competence for career concerns, politicians for re-election
purposes. The second assumption is that the tasks for bureau-cratic
agencies have to be specified ex ante and cannot be contingent on
the realization of too many shocks on the environment or on the pub-lic’s
preferences. If one accepts these two hypotheses, the nature of our
results is quite robust to variations on other less important assumptions.
>From a normative perspective, these differences between bureau-crats
and politicians imply that some policy tasks, but not others, ought
to be delegated to independent agencies. Consider first policies with few
redistributive implications, such as monetary policy or foreign policy.
Bureaucrats are likely to be better than politicians if the criteria for
49

good performance can be easily described ex-ante and are stable over
time; if good performance requires special abilities that wary widely in
the population and performance evaluation presupposes some technical
expertise; if political incentives are distorted by time inconsistency or
short-termism. Monetary policy indeed fulfills many of these conditions,
and the practice of delegating it to an independent agency accords with
some of these normative results. Foreign policy does not, because the cri-teria
for good performance are unstable and more vague, and the benefit
of insulating policy from the political process are smaller.
Next, consider policies that have redistributive implications, such
as trade policy, regulation, or fiscal policy. Here, bureaucrats perform
well if the policy consequences touch narrowly defined interest groups,
if criteria of good performance can be easily formulated and assessed in
terms of efficiency, and if the legal system is strong. Politicians instead
are better if the policy has far reaching redistributive implications so
that compensation of losers is important, if criteria of aggregate effi-ciency
do not easily pin down the optimal policy, and if there are in-teractions
across different policy domains (so that a single measure of
performance is affected by several policy instruments and policy pack-aging
is required to build consensus or achieve efficiency). Regulation of
public utilities or of specific industries are examples of policies that lend
50

themselves to bureaucratic delegation, since they pit special interests
against those of consumers as a whole, do not have large spillover ef-fects,
and policy performance can be evaluated on the basis of efficiency
or other semi-technical criteria. Trade policy might fall in this category
too, although here the redistributive implications are more pronounced.
Welfare state policies, instead, have such broad redistributive implica-tions
that it seems risky to subtract them from the political process, as
suggested by our examples on compensation of losers and cake splitting.
But there are specific aspects of fiscal policy that would certainly meet
our normative criteria for bureaucratic delegation: for instance, detailed
tax policy provisions, or intertemporal fiscal policy choices where time
inconsistency or political myopia is an obvious issue, as suggested by
Blinder (1997).
Overall, the normative analysis suggests that there is ample scope
for bureaucratic delegation to improve over political delegation, particu-larly
if politicians remain in charge of defining and correcting the general
mission of independent agencies. Are these normative conclusions likely
to be reflected in observed institutional arrangements? We explore this
positive question in a companion paper, Alesina and Tabellini (2004).
There we show that opportunistic politicians do not internalize these
normative criteria. Actual institutions are more likely to be designed so
51

as to deliver maximal rents at the lowest risk for the incumbent politi-cian.
This argues for retaining under political control policy tools that
are useful to build winning coalitions or to generate campaign contribu-tions,
such as trade policy or much of fiscal policy. It also means that
politicians might want to get rid of tasks that expose them to risk, such
as monetary policy. But this ”risk shielding” requires that bureaucratic
delegation be complete, so that the blame for policy failure lies fully
with the independent agency and does not reach the politician. This
might explain why it is politically so difficult to exploit delegation to in-dependent
agencies in fiscal policy. Full bureaucratic delegation of fiscal
policy is inconceivable, for normative and positive reasons. But par-tial
delegation of narrowly defined technical tasks in fiscal policy may
be politically unfeasible, no matter how desirable. The reason is that
voters would still hold the politician accountable, as long as he retains
some control (i.e. unless the delegation is complete). And if he is held
responsible, then the politician loses any incentive to delegate control.
Appendix
1. The optimal contract
Consider the simple model of section 2. If effort a is verifiable and
contractible, then the optimal contract induces the first best level of
effort, a∗, defined implicitly by:
52

1 = Ca(a∗) (36a)
Next, suppose that effort is unobservable, but performance y is verifiable
and contractible. Given risk neutrality of principal and agent, the first
best can still be achieved by an optimal explicit contract rewarding the
agent with a simple linear payoff based on performance:
R(y) = y − w
where the constant w is defined by the agent’s (ex-ante) participation
constraint, namely by the condition that
E(R(y)) − C(a) ≥ 0 (37)
Under the optimal performance contract, the participation constraint
must bind, and given (1) and (37), this implies: w = ¯θ
+ a∗ − C(a∗).
2. More general objective function for the
politician.
As mentioned in section 2, the politican’s objective function could be
written more generally by assuming tha he cares about both re-election
and, conditional on losing office, his career prospects outside politics. In
this case, his reward function could be written as:
RP (a) + P (W − a)RB(a) = β[1 − P (W − a)] + P (W − a)αE(y − ae)
53

where as before P (.) is the probability of losing the election. The first
order conditions for effort evaluated at the equilibrium are:
n(¯θ
)(β − α¯θ
) +
1
2
α = Ca(aP )
If the value of political office is sufficiently high compared to the expected
benefit of a career outside politics (if β is sufficiently higher than α¯θ
),
then the main implication of our model would still hold.
3. Lobbying
As stated in the text, the equilibrium with campaign contributions
must solve the following optimization problem by choice of a1,a2 and f,
subject to non-negativity constraints on the three choice variables, and
taking voters’ expectations ae
1 as given.
Max [(1 + γ)a2 − f ] s. to Pr(θ ≥ ¯θ
1 −a1 −H(f )) −C(a1+a2) ≥ ¯W
+ae
(38)
where ¯W
= Pr(θ ≥ ¯θ
1 − â1) − C(â1) is the politician’s utility if
+ ae
he refuses the campaign contributions and unexpectedly devotes effort
to please the voters (given that voters’ expectations ae
1 are consistent
with the equilibrium outcome). The out-of-equilibrium level of effort â1
is defined implicitly by the optimality condition: Ca(â1) = n(¯θ
+ ae
1 −
â1), and also depends on voters expctations of the equilibrium outcome.
In equilibrium, voters expectations must be consistent with the task
54

allocation chosen by the politician.
Letting λ denote the Lagrange multiplier of the constraint that the
politician is indifferent betwen acceptring or refusing the campaign con-tributions,
the optimality condition of the lobby’s optimization problem
imply:
n(¯θ
− H(f )) − Ca(a1 + a2) ≤ 0 (39)
1 + γ − λCa(a1 + a2) ≤ 0 (40)
λn(¯θ
− H(f ))Hf (f ) − 1 ≤ 0 (41)
where a strict inequality implies respectively: a1 = 0, a2 = 0, f = 0.
Consider first the case Hf (0) < 1/(1 + γ). Since Hf f < 0, lobbying
is inefficient and the first order conditions can only be satisfied if f =
a2 = 0 and a1 is at an interior optimum defined by n(¯θ
) − Ca(a1) = 0.
Next, consider the case Hf (f ∗) > 1/(1 + γ). This is the opposite
extreme, in which lobbying is very effective. In this case a1 = 0 and a2
and f are at an interior optimum defined jointly by
Ca(aP2
) = (1+γ)Hf (f ∗)n(¯θ
− H(f ∗))
and by the politician’s indifference condition (with ¯W
evaluated at the
1 = 0), namely:
point ae
Pr(θ ≥ ¯θ
− H(f ∗)) − C(aP2
) = ¯W
(42)
55

In the intermediate case, in which Hf (0) > 1/(1+γ) but the returns
to campaign contributions fall rapidly, we could also have an equilibrium
with positive campaign contributions but where the politicians devotes
effort to both tasks. In this case the equilibrium outcome is defined
implictly by the politicians’s indifference condition (42), and by the op-timality
conditions evaluated at an interior optimum for all three choice
variables, which implies:
(1 + γ)Hf (f ∗)=1
n(¯θ
− H(f ))=Ca(aP1
+ aP2
)
In the last two cases, the lobbymust also be better off than in the absence
of campaign contributions, i.e. (1 + γ)aP2
≥ f ∗. Voters are always made
worse off by positive campaign contributions, since they reduce effort in
the preferred task a1.
References
[1] Aghion, Philippe, Alberto Alesina, and Francesco Trebbi (2004)
”Endogenous Political Institutions” Quarterly Journal of Eco-nomics,
119, 565- 612.
Aghion, Philippe, Alberto Alesina, and Francesco Trebbi (2005)
”Choosing Electoral Rules: Theory and Evidence from US Cities”
NBER working paper no. 11236 (March).
Aghion, Philippe and Patrick Bolton (2003) "Incomplete Social
56

Contracts" Journal of the European Economic Association, 1, 38-
67.
Alesina A. and G. Tabellini (2004) ”Bureaucrats or Politicians?”,
Harvard University and Bocconi University, mimeo.
Besley T. and S. Coate (2003), "Elected versus Appointed Regu-lators:
Theory and Evidence", Journal of the European Economic
Association, forthcoming.
Besley T. and M. Ghatak (2003)” Competition and Incentives with
motivated agents” unpublished.
Blinder A. (1997) Is government too political? Foreign Affairs
Business Council of Australia (1999) "Avoiding boom/bust: macro
economic reforms for a globalized economy" Disc. Paper, Oct.
Carpenter D. (2001) Reputations, Negotiations, and Policy Innova-tion
in Executive Agencies 1862-1928 Princeton University Press,
Princeton NJ
Dewatripont, M., Jewitt, I., and Tirole, J. (1999a). ”The economics
of career concerns, part I: Comparing information structures.” Re-view
of Economic Studies 66, 183—198.
Dewatripont, M, Jewitt, I. and Tirole, J. (1999b). ”The economics
of career concerns, part II: Application to missions and account-ability
of government agencies", Review of Economic Studies 66,
57

199—217.
Drazen, D. and M. Eslava (2004) "Political budget cycles with-out
deficits: expenditure composition effects", mimeo, University
of Maryland
Epstein D. and S. O’ Halloran (1999) Delegating Powers Cambridge
University Press, Cambridge UK
Ferejohn, J. 1986. ”Incumbent performance and electoral control.”
Public Choice 50: 5-26.
Glaeser, E. and A. Shleifer (2003) "The rise of the Regulatory
State", Journal of Economic Literature
Grossman E. and G. Helpman (2001) Special Interest Politics MIT
press Cambridge Mass.
Hart O., A Shleifer, and R. Vishny (1997) ”The proper scope of
government: theory and application to prisons” Quarterly Journal
of Economics, 1127-61
Holmstrom B. and P. Milgrom (1991) ”Multitask Principal Agent
analysis: Incentive contracts asset ownership, and job design” Jour-nal
of Law and Economics, 24-52.
Maskin, E. (2001) "On Indescribable Contingencies and Incomplete
Contracts", mimeo, Princeton University, Institute for Advanced
Studies
58

Maskin E. and J. Tirole (2001) ”The judges and the politicians”
unpublished.
Persson T. and G. Tabellini (2000) Political Economics. Explaining
Economic Policy, MIT Press.
Rogoff, K. (1985). ”The optimal degree of commitment to an in-termediate
monetary target.” Quarterly Journal of Economics 100:
1169-1190.
Seabright, P. 1996. ”Accountability and decentralization in govern-ment:
An incomplete contracts model.” European Economic Review
40: 61-89.
Schultz, C. (2003) "Information, polarization and delegation in
democracy", University of Copenhagen, mimeo
Wilson J.Q. (1989) Bureaucracy Basic Books, New York.
59

Bureaucrats or politicians

More Related Content

What's hot

Viewers also liked

Similar to Bureaucrats or politicians

Recently uploaded

Bureaucrats or politicians