EC 134: Topics in Applied Economics (1a)
Week 6: Political Business Cycles, Deficit Spending and
Strategic Budgeteering
What is a political business cycle?
What is a political business cycle?
All else equal, politicians prefer winning elections to losing them.
All esle equal, voters prefer not to be fooled.
Incumenbents aim to increase the voters’ welfare before elections (particularly by
raising economic growth, by providing more public goods, or giving pre-electoral gifts).
If governments act purely opportunistic and if either voters’ memory does not last
forever or voters consider current conditions more than previous conditions, then the
incumbent is tempted to use all available policy instruments to create a business cycle
hike before elections at the cost of worsening economic conditions shortly after the
elections.
Can governments use monetary or fiscal policies to achieve this goal?
Monetary Cycles – the Phillips Curve
Phillip’s Curve:
William Phillips: 1958 “The Relation between Unemployment and the Rate of Change
of Money Wage Rates in the United Kingdom, 1861-1957”. In 1960 Paul Samuelson
and Robert Solow took Phillips' work and made explicit the link between inflation and
unemployment: when inflation was high, unemployment was low, and vice-versa.
Nordhaus (1975) transformed the Phillip’s Curve argument into a political business
cycle theory.
retrospective or rational forward looking voters? (Alesina, Rogoff and Sibert, Persson
and Tabellini)
1970s – stagflation: Longterm Phillips Curve – NAIRU: the inflation target of the
government equals the natural output growth, natural unemployment rate
-
Competency and partisan ideology
Time inconsistency, credible committment – CBI, fixed ER, CU
Monetary policy autonomy?
weak empirical evidence at best
Political Budget Cycles (PBC)
PBCs are cycles in some component of the government budget induced by the
electoral cycle.
Ususally: increases in government spending/deficit or tax cuts in an election year
because incumbent wants to be re-elected
2 Main Views on PBCs:
1. Voters expect Politicians to engage in such manipulation
Voters like low taxes and high government spending and vote for incumbents who like
them
Opportunistic incumbents will thus use expansionary fiscal policy before elections to
increase the probability of re-election
Inconsistent with rational forward looking voters – why?
Rational, forward-looking voters are aware of budget constraints both at a point in time
in intertemporally
Non-smooth paths of taxes and spending (election-year deficits) = costly  voters
dislike electorally motivated deficits
2. Voters are “fiscal concervatives”
Voters punish rather than reward fiscal manipulation, but they respond positively to
good economic conditions
Voters have imperfect information about the ability of the candidate or the environment
 fiscal expansion signals incumbent’s competence (Rogoff 1990)
More competent incumbents can deliver more public goods at same level of taxes –
since competence is correlated over time, these politicians are also more competent
after elections
The Rogoff Model
2 periods, election after first period, rational voters, voters choose leader on the basis
of any information they gather in the first period, politicians are both office and welfare
motivated (separating equilibrium) and differ in their ability (low and high types)
U of voter = f(public good, public investment, random shock (looks))
Incumbent has a level of competence to produce the public good which cannot be
observed before the election
Investment can only be observed in t+1 (after election)  if voter observes high PG in t,
she cannot infer whether due to competence or at expense of investment (basic
inference problem)  vote choice based on observation of PG
Given the voting rule, incumbent has an incentive to appear to be of high ability
High types of leaders can increase PG at lower cost (less cut back in investment) 
signal less costly for high type.
Separating equilibrium – level of PG reveals incumbents competence type: high type
manipulates, low type chooses first best solution and loses election with certainty  on
average we observe a budget cycle (in public good spending/ deficit)
Implication: higher the degree of transparency, the lower the distortion way from the
first best policy in the PBC
Extensions:
Shi and Svensson (2002): policymaker chooses fiscal policy before she knows her
competence level  all types choose the same level of fiscal expansion (moral
hazard), same if candidates are only office motivated  pooling equilibrium
Drazen and Eslava (2005/2006): incumbent’s preferences = unobserved  change of
composition of spending (targeted spending)  inference problem for voter, whether
targeted spending is electorally motivated or persistent (politician’s preference)
Political Budget Cycles

Pre-electoral deficit spending (Rogoff and Sibert, Shi And Svensson, Franzese,
Cusack, Boix etc.) – some evidence

Pre-electoral tax cuts – weak evidence

Availability of information (transparency) (Alt and Lassen)

Targeted spending (swing districts) (Drazen and Eslava)
Summary of Government Constraints and Policy Options
central bank independence
low
high
exchange rate peg
no
yes
'fear of floating'
low
high
fiscal transparency
low
monetary policy
deficit spending
high
strategic budgeteering
Strategic Budgeteering
All else equal, politicians prefer winning elections to losing them.
All else equal, voters prefer not to be fooled.
Incumbents aim to increase the voters’ welfare before elections (i.e., by providing more
public goods, or giving pre-electoral gifts).
Can governments use fiscal policies to achieve this goal?
i)
ii)
iii)
deficit spending (Shi & Svensson 2002; Drazen 2000, Persson & Tabellini 2003)
tax revenues (Persson & Tabellini 2002/3)
compositional, targeted spending (Rogoff 1990; Drazen & Eslava 2005/2006; Katsimi &
Sarantides 2012, Brender & Drazen 2013, Morozumi, Veiga & Veiga 2014)
PBC and Deficit Spending
Empirical evidence is weak at best:
Ability to create political budget cycles (deficit spending) depends on availability of
information that voters and opposition can obtain
i)
ii)
i)
fiscal transparency (Alt & Lassen 2006, Alt & Lowry 2010, Alt, Lassen & Wehner 2014)
state-enforced media (Ferejohn 1999, Akhmedov and Zhuravskaya 2004)
new and weak democracies (Persson and Tabellini 2002/2003, Brender and Drazen 2005; Shi
and Svensson 2006; Brender & Drazen 2013, Morozumi, Veiga & Veiga 2014)
Main Question: Has fiscal transparency eliminated the ability of governments in
developed democratic countries to pursue opportunistic fiscal policies before elections?
How do governments choose between alternative fiscal policy instruments, given the
(domestic/international) institutional constraints they face.

Politics of strategic budgeteering
The Argument
Voters want to maximize their expected income and value increased consumption.
But they are fiscal conservatives (fully rational): Voters prefer candidates who are able to
provide more public goods for given levels of taxation and without increasing the deficit
Incumbents want to signal economic competence, potentially through hidden action
(Rogoff 1990 – signalling competence; Shi and Svensson 2002 – moral hazard).
Voters have (temporarily) imperfect information about the incumbent’s ability 
Availability of Information affects basic inference problem for voters: Fiscal
transparency allows voters and opposition to observe public deficits. (Besley: screening
and discipline)
The Argument (cont.)
 Opportunistic governments want to be re-elected
 They can choose
 long-term efficient public investments (only observed by voters after election)
 short-term beneficial public consumption
 governments always have an incentive to increase public good provision
before elections
 Incentive increases when election is close
 They can do so by
 increase deficits (with re-balancing the budget after elections)
 redirect resources from (long-term efficient) public investments to shortterm
consumption (compositional spending)
 They are restricted by
 budget constraints
 fiscal transparency
Predictions
Deficit spending is more flexible than compositional spending, but not feasible in fiscally
transparent economies.
Compositional spending is less flexible, but more feasible in transparent economies.
 Trade off between flexibility and feasibility.
H1: The lower fiscal transparency, the greater the likelihood that governments use
deficit spending to increase public goods provision in preelection periods, ceteris
paribus.
H2: The greater fiscal transparency, the more likely governments use compositional
spending to increase public goods provision in pre-election periods, ceteris paribus.
Other implications:
 Closeness of elections
 Interest rates
 EMU, EU membership
What are we talking about?
Ottawa — The Globe and Mail: 27/07/2015, headlines:
“Conservatives hiding budget details ahead of election, opposition says”
“The gifts that the Conservatives are throwing around on the eve of an election right
now – the cheques that they are handing out – if deficit-financed, will actually be paid
for in the future by cuts to services or higher taxes,” Scott Brison, the Liberal finance
critic, told reporters
Bruce Bartlett, Forbes: about Reagan’s Medicare drug benefit:
“[…] But a big election was coming up that Bush and his party were desperately fearful
of losing. So they decided to win it by buying the votes of America’s seniors by giving
them an expensive new program to pay for their prescription drugs.”
Empirical Analysis I
 32 (mainly OECD) countries (1970-2012)
 Main DVs: Change in Government Debt, Public Consumption, Ratio of government
consumption to investment spending
 Main IVs:
- Preelectoral Period (dummy, m /12 in election year 

1  m /12 year before election)

 tripleinteraction effect
- Change in government debt

- Fiscal transparency  OBI and Alt / Lassen index  
 Other explanatory variables:
- Closeness of election
- Interest rates
- EMU membership
 Controls: Unemployment, pc GDP, government control, fractionalization, checks
and balances, polarization, electoral system, FDI, GDP pc growth, gross savings,
trade, inflation, age dependency ratio, tax revenue, etc
 Main model specification: Fixed effects TSCS regressions, trends, clustered SEs
Robustness Checks
Alternative dependent variables (consumption vs. investment vs deficit/debt; individual
items)
Alternative fiscal strategies (interest rates, inflation, taxation)
Alternative model specifications (endogenous elections, random effects, etc)
Post-election effects
Alternative Samples: Established vs weak democracies
Partisan cycles
Fiscal Transparency
Has come into the focus of IOs during the last two decades and has improved greatly
but not uniformly since then:
 IMF Fiscal Transparency Code: first introduced in 1998, revised in 2007, provides a
comprehensive framework for fiscal transparency and focuses on clear roles and
responsibilities, transparent budget processes, public availability of information, and
assurances of integrity.
 OECD Best Practices for Budget Transparency: issued in 2001 and are to be used as a
reference tool. They support the full disclosure of all relevant fiscal information in a timely and
systematic manner and provide a series of best practices in the areas of principal budget
reports, specific disclosures, quality, and integrity.
 Open Budget Initiative: The Open Budget Index (2005) provides ratings of the openness of
budget material in 59 countries based on expert surveys. Assessment of availability of key
budget documents, the quantity of information they provide, and the timeliness of their
dissemination to citizens.
 EU: acquis communitaire, Maastricht criteria and the Stability and Growth Pact
Fiscal Transparency Measures
Mexico
Indonesia
Turkey
Slovak Republic
Portugal
Japan
Italy
Russia
Czech Republic
Spain
Belgium
India
Ireland
Germany
Iceland
Switzerland
Canada
Korea
Finland
Denmark
Poland
Brazil
Netherlands
Australia
Austria
Slovenia
United States
Sweden
Norway
United Kingdom
France
New Zealand
Japan
Germany
Norway
Switzerland
Belgium
Denmark
Ireland
Italy
Austria
Canada
Finland
France
Iceland
Sweden
Netherlands
Australia
United Kingdom
United States
New Zealand
40
50
60
70
80
fiscal transparency (open budget index)
90
0
1
2
3
4
5
6
7
8
fiscal transparency (Alt/Lassen)
9
10
11
“public openness about the structure and functions of government, fiscal policy intentions, public
sector accounts, and projections. It involves ready access to reliable, comprehensive, timely,
understandable, and internationally comparable information on government activities (…) so that
the electorate and financial markets can accurately assess the government’s financial position and
the true costs and benefits of government activities, including their present and future economic
and social implications” (Kopits and Craig 1998, 1).
Main Empirical Results
Pre-electoral effects on alternative measures to deficit spending and budgeteering:
interest rates, inflation, taxation  no effects
Pre-electoral Effect on Change in Debt (+ controls)  positive pre-election effect
damped by fiscal transparency
Pre-electoral effects of changes in debt and transparency on consumption per GDP 
deficit spending is used to increase pre-electoral public consumption if transparency is low,
no effect if transparency is high
Pre-electoral effects of changes in debt and transparency on Government
consumption/Investment Ratio  deficit spending is used to increase pre-electoral public
consumption if transparency is low, and investment is used if transparency increases
Empirical Results
Pre-electoral effects on alternative measures to deficit spending and budgeteering
DV:
Pre-election period
Inflation
Real
interest
rate
-2.620
-1.968
(21.866) (2.040)
-0.074
-0.030
Fiscal Transparency
(OBI)
(0.302)
(0.029)
Preelection*transparency -0.044
0.046
(0.371)
(0.034)
Intercept
32.179** 7.283***
(15.928) (1.587)
2
R (within)
0.000
0.003
N
1235.000 949.000
Number of countries
37.000
35.000
F statistic
0.111
0.806
Tax
revenue
Statutory
corporate
tax
0.132
(0.470)
0.000
(0.004)
-0.003
(0.007)
18.254***
(0.265)
0.001
547.000
36.000
0.182
0.007
(0.014)
-0.001***
(0.000)
-0.000
(0.000)
0.447***
(0.011)
0.072
468.000
18.000
11.583
Effective
average
corporate
tax
0.001
(0.010)
-0.001***
(0.000)
0.000
(0.000)
0.323***
(0.007)
0.050
468.000
18.000
7.763
Average
income tax
0.156
(0.585)
-0.017**
(0.007)
-0.001
(0.010)
18.068***
(0.414)
0.008
821.000
29.000
2.185
Pre-electoral Effect on Change in Debt (+ controls)
DV: ∆ debt
Pre-election period
Fiscal Transparency
(OBI/Alt-Lassen)
Preelection*transparency
EMU membership
Preelection*EMU
Endogenous election
timing (1=yes)
Preelection*endog elect
Margin of Majority
Preelection*Margin of
Majority
OBI
0.818
(0.936)
-0.020*
(0.012)
-0.008
(0.016)
OBI
2.089
(2.647)
-0.014
(0.013)
-0.009
(0.015)
-2.252***
(0.825)
0.568
(1.105)
2.294*
(1.297)
1.078
(1.016)
-1.236
(4.831)
-5.164
(4.141)
Alt/Lassen
2.691**
(1.065)
Dropped due to
collinearity
-0.431*
(0.230)
Alt/Lassen
2.233
(2.729)
Dropped due to
collinearity
-0.373*
(0.194)
-1.932**
(0.824)
-0.603
(1.124)
2.577**
(1.175)
-0.193
(1.090)
-6.280
(4.940)
-0.391
(4.256)
Pre-electoral Effect on Change in Debt
0
-5
pre-election effect
5
pre-elector
0
5
fiscal transparency Alt/Lassen
10
Pre-electoral effects of changes in debt and transparency on consumption per GDP
DV:
∆ debt
Pre-election period
Fiscal Transparency
(OBI)
Preelection*transparency
Preelection*debt
Debt*transparency
Peelection*debt*trans
EMU membership
Preelection*EMU
Preelection*debt*EMU
Endogenous election timing (1=yes)
Preelection*endog elect
Margin of Majority
Preelection*Margin of Majority
Consumption
0.058***
(0.018)
-1.297
(0.853)
-0.023
(0.014)
0.020**
(0.008)
0.020**
(0.008)
0.001***
(0.000)
-0.000*
(0.000)
consumption
0.050***
(0.010)
-1.246
(0.848)
-0.025***
(0.005)
0.012
(0.009)
0.012*
(0.007)
0.000***
(0.000)
-0.000
(0.000)
0.245
(0.210)
-0.094
(0.764)
0.004
(0.010)
0.120
(0.329)
0.045
(0.261)
-4.009***
(1.001)
0.431
(1.051)
.015
0
.005
.01
Kernel Density Estimate of OBI
0
.02
-.02
marg effect of preelectoral deficit
.02
pre-electoral effect of deficit spending on consumption
.04
Pre-electoral effect on government consumption
0
20
40
60
fiscal transparency
80
100
Government consumption/Investment Ratio
DV: consumption/investment
∆ debt
Pre-election period
Fiscal Transparency
(OBI/Alt-Lassen)
Preelection*transparency
Preelection*debt
Debt*transparency
Peelection*debt*trans
Trend
EMU membership
Preelection*EMU
Preelection*debt*EMU
Endogenous election timing (1=yes)
Preelection*endog elect
Margin of Majority
Preelection*Margin of Majority
Baseline
OBI
0.006***
(0.002)
-0.116**
(0.052)
-0.004***
(0.001)
0.002**
(0.001)
0.002**
(0.001)
0.000***
(0.000)
-0.000**
(0.000)
0.006***
(0.002)
Baseline
Alt/Lassen
0.007***
(0.002)
-0.108***
(0.034)
Dropped due to
Collinearity
0.034**
(0.012)
0.002***
(0.000)
0.001**
(0.000)
-0.001**
(0.000)
0.005***
(0.002)
extended
OBI
0.007***
(0.001)
-0.096
(0.072)
-0.003***
(0.000)
0.002**
(0.001)
0.001**
(0.001)
0.000***
(0.000)
-0.000**
(0.000)
0.007***
(0.001)
-0.032*
(0.018)
-0.017
(0.066)
0.000
(0.001)
0.030
(0.028)
-0.011
(0.022)
-0.099*
(0.056)
-0.001
(0.089)
Pol. controls
OBI
0.007***
(0.001)
-0.131
(0.081)
-0.004***
(0.000)
0.002**
(0.001)
0.001**
(0.001)
0.000***
(0.000)
-0.000**
(0.000)
0.007***
(0.001)
-0.026
(0.019)
-0.035
(0.069)
0.001
(0.001)
0.026
(0.029)
-0.001
(0.025)
-0.346***
(0.098)
0.035
(0.105)
full
OBI
0.002
(0.001)
-0.238*
(0.127)
-0.003***
(0.001)
0.003*
(0.002)
0.002
(0.002)
0.000***
(0.000)
-0.000
(0.000)
-0.002
(0.002)
-0.075***
(0.026)
-0.008
(0.063)
0.000
(0.001)
Dropped due to
collinearity
0.007
(0.028)
-0.047
(0.123)
0.139
(0.114)
0
5
pre-election
10
15
.01
.012
.005
-.002
-.004
.002
.15
0
20
40
60
fiscal transparency
.01
.015
Kernel Density Estimate of OBI
0
.1
marg effect of preelectoral deficit
.05
0
.008
Kernel Density Estimate of pre12_legelec
.006
0
.004
.02
.004
.2
.014
Effect of deficit spending when elections get
closer
Preelectoral Effect of Deficit Spending on
Government/Investment Ratio
80
100
Other Findings
Closeness of elections strengthen political budget cycles
No preelectoral effects for revenues or monetary policies
Small postelectoral effect for tax revenues (increase)
No effect of EMU membership on consumption-investment ratio (but overall better fiscal
performance)
No effect of endogenous election timing
Partisanship matters
Other Explanations and Measures
State Controlled Media, Media Diffusion (Morozumi, Veiga & Veiga 2014, Ferejohn 1999)
Switzerland
Denmark
Belgium
Iceland
Norway
New Zealand
Luxembourg
Sweden
Netherlands
United States
Australia
Poland
Estonia
Germany
Ireland
Canada
Finland
Portugal
Czech Republic
Slovak Republic
Spain
United Kingdom
Slovenia
Austria
France
Japan
Hungary
Italy
Greece
Korea
Brazil
India
Turkey
Indonesia
Mexico
Russia
0
10
20
plitical control over media
30
Mature Democracies (Brender & Drazen 2013, Morozumi, Veiga & Veiga 2014)
Indonesia
Mexico
Russia
Korea
Greece
Turkey
Portugal
Brazil
Estonia
France
India
Czech Republic
Spain
Slovak Republic
Belgium
Australia
Austria
Canada
Denmark
Finland
Germany
Hungary
Ireland
Italy
Japan
Luxembourg
Netherlands
New Zealand
Norway
Poland
Slovenia
Sweden
Switzerland
United Kingdom
United States
Mexico
Estonia
Korea
Slovenia
Russia
Czech Republic
Poland
Hungary
Indonesia
Slovak Republic
Greece
Spain
Portugal
Germany
France
Brazil
Turkey
Italy
Japan
Netherlands
Finland
Austria
Norway
Belgium
Denmark
India
Luxembourg
Ireland
Sweden
Canada
Australia
United Kingdom
New Zealand
Switzerland
United States
2
OBI
Tran AL
media
Durability
Regime type
4
10
0
Open budget Transparency Pol. control
index
AL
of media
Durability
0.42
0.01
0.25
-0.25
6
regime type
-0.19
0.54
0.03
8
-0.28
-0.19
0.24
50
100
durability of democracy
150
200
0
10
20
30
political control over media
40
50
0
50
100
durability of democracy
150
200
0
-.0005
.04
.06
0
.08
-10
0
.01
.0005
-5
.1
.2
.3
.4
Kernel Density Estimate of polity2
-.0005
.005
marg effect of preelectoral deficit
Kernel Density Estimate of durable
.02
marg effect of preelectoral deficit
.5
.001
.015
.0005
.1
.0005
Durability of Democracy
0
-.001
-.0005
Kernel Density Estimate of pol_media_freedom
-.001
Media Freedom
0
-.001
0
-.0015
Empirical Findings
Regime Type
0
regime type
5
10
Empirical Analysis II
 Composition of government spending: 17 OECD countries, 35 years (1970 -2004)
 1st stage: alternative to strategic budgeteering: deficit spending, monetary policy
(interest rate, inflation), taxation (consumption tax revenue, average effective labor
and capital tax)
 2nd stage: expenditure for 23 (14 allocated) spending items, relative and total
Explanatory Variables:






Lonterm efficient vs. Shorttem spending items
Pre-electoral Period (m/12, 1-m/12, pre-election year dummy)
Fiscal Transparency (0-11) (Alt & Lassen, OBI)
Closeness of election (margin of majority)
Partisan specific budget items
Controls: Unemployment, GDP p.c., left and right cabinet portfolios, electoral
system, checks and balances
Model Specification: fevd, P-W, direct test for substitution effects, SEM
Budget Composition
Expenditure
Item
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Description
General Public Services
Defense
Public Order & Safety
Education
Health
Social Security & Welfare
Housing & Community Amenities
Recreational, Cultural & Religious Affairs
Fuel & Energy
Agriculture, Forestry, Fishing & Hunting
Mining, Manufacturing & Construction
Transportation & Communication
Other Economic Affairs & Services
Other Expenditures
of which: Interest Payments
Adjustment to Total Expenditure
Economic Affairs & Social Services (9-13) (only if 9-13 not available)
Environmental Protection
Electricity, Steam, Water
Roads
Inland & Coastal Waterways
Other Transport & Communication
Other Economic Services
Longterm/ shortterm
Longterm
Longterm
Longterm
Longterm
Shortterm / non-targeted
Shortterm
Shortterm/ Targeted
Shortterm/ Targeted
Shortterm/ Targeted
Shortterm/ Targeted
Shortterm
Longterm
Shortterm
Shortterm
Empirical Results I
Pre-election
Pre*Transp.
Transparency
Unemployment
GDP per cap.
Left cabinet
Right cabinet
Checks & Bal.
Majoritarian
Interest Rate
Model 1a
Deficit
Model 1b
Inflation
Model 1c
Interest rate
4.901**
(2.394)
-0.695**
(0.363)
-0.197*
(0.121)
3.757***
(0.192)
0.001***
(0.000)
0.038**
(0.015)
0.014
(0.018)
-1.176***
(0.327)
-19.502***
(0.964)
-0.302**
0.207
(0.430)
0.004
(0.108)
-0.365***
(0.090)
-0.768***
(0.061)
-0.001***
(0.000)
-0.012***
(0.004)
-0.012***
(0.004)
-0.042
(0.051)
0.766***
(0.122)
-0.112
(0.470)
0.038
(0.121)
-0.206***
(0.073)
-0.211***
(0.066)
-0.000***
(0.000)
-0.006
(0.007)
-0.009
(0.007)
-0.089*
(0.059)
0.478**
(0.190)
Model 1d
Consumption
tax
28.059
(34.833)
-4.210
(6.872)
18.858***
(2.529)
0.204
(2.626)
0.026***
(0.002)
-0.511
(0.400)
-0.231
(0.488)
-11.256**
(5.612)
347.782***
(25.752)
Model 1e
Labor tax
Model 1f
Capital tax
-0.578
(0.531)
0.112
(0.103)
0.546***
(0.032)
0.647***
(0.052)
0.000***
(0.000)
0.012*
(0.006)
0.003
(0.007)
-0.036
(0.112)
-12.162***
(0.249)
-0.419
(0.959)
0.104
(0.209)
0.108
(0.106)
-0.085
(0.115)
0.000
(0.000)
0.008
(0.014)
0.010
(0.010)
-0.126
(0.164)
8.913***
(0.409)
(0.128)
EMU
-2.207**
(0.945)
Intercept
12.810***
26.998***
18.968***
-324.723***
(0.449)
(0.294)
(0.269)
(18.732)
Fevd
fevd
Fevd
fevd
R² adj.
0.878
0.644
0.561
0.897
N
439
477
426
480
F
14595.300
707.237
42.018
184.135
Prob. > F
0.000
0.000
0.000
0.000
Clustered Standard Errors in Parentheses; *** p<0.01, ** p<0.05, * p<0.1
28.365***
(0.289)
Fevd
0.909
492
1429.866
0.000
22.831***
(0.395)
fevd
0.786
485
395.294
0.000
Empirical Results: Strategic Budgeteering
Model 2a
Relative
exp. per
item
Total gov. exp.
Preelection longterm
spending
Pre*long*Transp.
Preelection shortterm
spending
Pre*short*Transp.
Pre*long*debt
Pre*short*debt
Model 2b
Total exp. per
item
-0.04***
0.088***
(0.006)
-3369.75***
(0.010)
0.004**
(0.002)
0.043**
(0.017)
-0.003
(0.003)
Model 2c
Relative
exp. per
item
Model 2d
Total exp. per
item
Model 2e
Relative
exp. per
item
Model 2f
Total exp. per
item
-0.02***
0.091***
(0.008)
-3808.30***
-0.022**
0.091***
(0.008)
-2394.334*
(933.888)
290.745**
(123.602)
5654.359**
(0.006)
0.002*
(0.001)
0.057***
(1061.337)
343.850**
(139.392)
6567.498**
(0.009)
0.002*
(0.001)
0.048*
(1263.005)
267.070**
(134.216)
5304.239*
(2454.492)
-608.965*
(321.497)
(0.016)
-0.005*
(0.003)
(2856.804)
-734.906**
(373.468)
(0.024)
-0.004*
(0.003)
-0.000
(0.000)
0.000
(0.000)
(3552.546)
-665.477*
(382.674)
-24.356
(18.933)
22.458
(50.112)
Debt per GDP
0.000
14.940
(0.000)
(21.620)
Inflation
0.000
75.165**
(0.000)
(31.427)
Labor tax
0.001
18.970
(0.000)
(61.560)
Capital tax
-0.000
74.479*
(0.000)
(44.694)
Consumption tax rev.
0.000
-0.091
(0.000)
(0.826)
Intercept
0.072***
-1378.037*
0.061***
-5440.959**
(0.004)
(739.683)
(0.016)
(2666.214)
Country dummies
Yes
yes
Yes
Yes
R² adj.
0.004
0.283
0.009
0.278
N
6268
6268
5801
5801
F
4.106
79.183
3.073
62.387
Clustered Standard Errors in Parentheses; *** p<0.01, ** p<0.05, * p<0.1
0.000
(0.000)
0.000
(0.000)
0.001
(0.000)
-0.000
(0.000)
0.000
(0.000)
0.061***
(0.016)
Yes
0.009
5801
2.949
15.916
(22.294)
75.641**
(31.492)
17.573
(61.696)
74.523*
(44.688)
-0.071
(0.828)
-5429.18**
(2666.167)
yes
0.278
5801
57.754
Main Empirical Results
2
4
6
Fiscal Transparency
8
10
0
0
.02
.04
.06
pre-election effect on relative shortterm spending
.02
Mean of Transparency
-.02
0
-.02
-.04
-.06
pre-election effect on relative longterm spending
8
6
4
2
0
-2
-4
-6
0
Mean of Transparency
Pre-elect. Short-term Consumption
.08
Pre-electoral Long-term Investment
10
Pre-electoral Deficit Spending
2
4
6
Fiscal Transparency
8
10
Mean of Transparency
0
2
4
6
Fiscal Transparency
8
10
Empirical Results: Strategic Budgeteering - extended
Model 3a
Relative exp.
Model 3c
Relative exp.
-0.031**
Model 3b
Total exp.
0.090***
(0.007)
-6532.98**
(0.015)
0.003*
(0.002)
0.006
(0.023)
0.039*
(0.026)
-0.004*
(0.003)
-0.017
(0.061)
Total gov. exp.
Preelection longterm
spending
Pre*long*Transp.
Pre*long*Margin maj.
Preelection shortterm
spending
Pre*short*Transp.
Pre*short*Margin
Inflation
Labor tax
Capital tax
Consump. tax rev.
Model 3e
Relative exp.
-0.020*
Model 3d
Total exp.
0.093***
(0.009)
-6033.321**
-0.019*
Model 3f
Total exp.
0.102***
(0.010)
-6738.809**
(2850.240)
280.040**
(125.950)
5810.986*
(4086.931)
10098.556*
(0.013)
0.002*
(0.001)
0.010
(0.023)
0.043
(2952.013)
346.340**
(140.816)
3907.761*
(2221.841)
11205.359*
(0.014)
0.001
(0.001)
0.015
(0.025)
0.027*
(2674.757)
372.816***
(129.299)
4695.976
(3048.595)
3630.990*
(7194.097)
-636.766*
(330.225)
-7487.568
(10161.018)
(0.036)
-0.005*
(0.003)
-0.028
(0.063)
0.000
(0.000)
0.001
(0.000)
-0.000
(0.000)
0.000
(0.000)
(7522.284)
-729.984*
(374.857)
-8321.073
(10429.683)
84.923**
(35.008)
8.118
(72.049)
85.560*
(52.075)
-0.388
(1.008)
(0.018)
-0.003*
(0.001)
-0.004
(0.031)
0.000
(0.001)
0.000
(0.001)
-0.000
(0.000)
-0.000
(0.000)
(2412.679)
-502.394*
(313.175)
-1.08e+04
(9827.533)
214.210*
(109.373)
-30.326
(102.137)
87.858*
(54.160)
2.523*
(1.785)
Checks & Balances
GDP per capita
Unemployment
Majoritarian
Margin of Majority
Left cabinet portf.
Right cabinet portf.
Pre*unemp*SoSec
Pre*Left*SoSec
Pre*Right*Def.
Intercept
0.073***
-1623.061*
0.062***
(0.005)
(861.843)
(0.016)
Country dummies
yes
yes
Yes
R² adj.
0.004
0.277
0.010
N
5742
5742
5391
F
4.003
68.301
3.008
Clustered Standard Errors in Parentheses; *** p<0.01, ** p<0.05, * p<0.1
-5915.837*
(3041.554)
yes
0.273
5391
55.134
0.000
(0.001)
0.000
(0.000)
-0.002*
(0.001)
0.034
(0.031)
0.014
(0.017)
-0.000
(0.000)
0.000
(0.000)
0.037***
(0.002)
0.002***
(0.000)
0.000***
(0.000)
0.030
(0.040)
Yes
0.172
4916
28.652
188.375
(187.133)
-0.302*
(0.182)
-83.586
(152.240)
23124.490**
(10130.891)
1383.456
(2413.600)
-3.783
(11.113)
-1.047
(9.808)
3554.101***
(516.566)
199.355**
(91.215)
12.898*
(9.051)
-2.60e+04**
(10842.177)
yes
0.325
4916
35.008
Robustness
 SEM for single budget items
 Regress short-term spending on long-term spending and vice versa
 Randomizing the assignment of budget items to short-term and long-term spending
– no significant results
Conclusions
 Governments always have an incentive to provide pre-electoral gifts
 To increase voters’ welfare before elections and thus boost the probability of
staying in office they choose between alternative fiscal strategies
 Deficit Spending is most attractive but less likely if fiscal transparency is high
 governments redirect resources from long-term investment to short-term
public good provision especially if elections are highly contested.
 Fiscal transparency constraints all incentives to manipulate the electoral
business cycle
 Partisan effects: incumbents redirect money to please partisan constituency
Extensions
 Extentions:
 partisan, regional, fuctional and sectoral budgeteering
 incorporating institutional constraints: electoral system – single party vs.
multi-party governments
 longterm vs. shortterm and non-targeted vs. targeted spending: concentration
of voter preferences
 Identification:
 Instances in EU regulation that changed de facto fiscal transparency for some
member states but not for others