Strenghtening Medium-Term

1

• Strenghtening Medium-Term
• Fiscal Frameworks

Fabrizio Balassone Econpubblica Università
Bocconi Milano, 25 Marzo 2009
2

• MTFF definition purpose

• definition set of institutions, procedures and
  rules governing (constraining) the development
  of public finances over the medium term
• purpose ensure fiscal discipline
  (sustainability stabilization)
• and efficient use of
  resources
• work on MTFF combines economics,
  institutional knowledge and
• a fair dose of pragmatism

3

• Fiscal discipline means...

• maintaining a prudent budget balance to
• ensure sustainability of public debt
• allow margins to face cyclical fluctuations and
  
• e unforeseeable events taking into account
  the
• degree of debt tolerance
• (Kumar Ter-Minassian, IMF 2007)

4

• Why budget discipline and efficiency?

• fiscal discipline and efficiency maximize public
  sector (PS)
• contribution to economic growth and welfare
• fiscal discipline prerequisite to PS functions
• budget constraint is nec. cond. for allocative
  efficiency
• low debt ? no financial fragility ? margins for
  stabilization
• efficiency stability ? growth ? resources for
  redistribution
• NB discipline does not imply efficiency need
  accountability
• budget transparency (informed and explicit
  choices)
• measurability of results (assessment)

5
Spreads (Spilimbergo et al. IMF09)
350
Long-Term Government Bond Yield Spread vis-à-vis
Germany
Greece
300
Ireland
250
200
Portugal
Basis points
Italy
150
Austria
Spain
Belgium
100
Finland
Netherlands
50
France
0
1/1/07
3/12/07
5/21/07
7/30/07
10/8/07
12/17/07
2/25/08
5/5/08
7/14/08
9/22/08
12/1/08
2/9/09
6

• The building blocks of a MTFF

1. a stable long-term anchor for the public finances
   rooted in some measure of sustainability
2. a simple medium-term rule ensuring consistency
   between the fiscal stance and the long-term
   anchor
3. a transparent convention on the headroom to build
   into the fiscal balances to deal with adverse
   circumstances
4. a multi-year budget with top-down preparation and
   tight execution procedures
5. mechanisms to ensure prudent forecasts of
   macroeconomic and fiscal variables
6. reporting, monitoring, auditing tools to ensure
   accountability

7

• OUTLINE

• Why discretion needs to be constrained
• Issues in building a MTFF (and solutions?)
• A few remarks about Italy
• Summary

8

• Why Does Discretion Need to Be Constrained?

9
The political economy of budget deficits

• opportunistic politicians naive voters
  (Puviani, 1903 Buchanan Wagner,
• AEI 77 Buchanan et al., 1986)
• variation 1 myopic politicians (Alesina
  Tabellini, RES 1990 Rogoff, AER 1990)
• variation 2 intergenerational distribution
  (Browning, EI 75
• Tabellini, NBER 90, JPE 91
• Cuckierman Meltzer, AER 89)
• time inconsistency (Kydland Prescott, JPE 77
  Eichengreen et al., OER 99)
• common pool (Eichengreen et al., OER 1999
  Velasco, 1999)
• variation 1 federalism (Buchanan, 1967
  Oates, 1979 Weingast et al. JPE81)
• variation 2 monetary union (Balassone
  Franco, JPFPC02)
• strategic use of the budget (Persson Svensson,
  QJE 89 Alesina Tabellini, RES 90
• Tabellini Alesina, AER 1990)
• coalitions wars of attrition (Roubini-Sachs,
  EP89 Alesina-Drazen, AER91
  Balassone-Giordano, PC01)

10
Some evidence of the propensity to deficit
finance
11
It gets worse in good times

• voracity effect perverse interaction between
  abundance of
• resources and political economy factors
  (common pool, myopia)
• (Lane Tornell, AER 1999 Debrun, Hauner
  Kumar, IMF 2007)
• growing supporting evidence (asymmetric
  procyclicality)
• (Buti et al. OREP98 Buti Sapir 1998
  Balassone Francese, TD04 TD08 European
  Commission, 2006)
• coming mostly from expenditure
• (Kaminsky et al., NBER04 Hercowitz and
  Strawczynski, RES04 Balassone, Francese,
  Zotteri, TD08)

12
Can we rely on markets?

• Theory
• (Bishop et al., 89 Lane, SP 93)
• No privileged access
• Full information
• No bail-out
• Timely Signals
• Sensitivity to signals

• Practice
• CB Independence
• Issue of information quality remains
• (Balassone, Franco, Zotteri, E06)
• Credibility of the clause? (IMF97)
• Not always (Ferri et al., EN99)
• Low and delayed
• (Balassone, Franco, Giordano, BI04)

Fitch Ratings (2004) 15-20 basis points is
perhaps the most that could be attributed to
credit differentials between AAA and AA euro-area
governments and such amounts hardly seem likely
to keep a German finance minister awake at night
(p. 6).
13
The quality of information
14
The quality of information
15
II. Issues in Building a MTFF
16

• The broad definition of fiscal discipline

• fiscal discipline prudent budget balance to
• ensure sustainability of public debt
• allow margins to face cyclical fluctuations and
  
• unforeseeable events taking into account
  the
• degree of debt tolerance
• Problems
• defining debt sustainability
• forecasting/measuring the economic cycle and its
  effects on the fiscal balance
• quantifying implications of unforeseeable events
• (e.g. contingent liabilities - IMF 2007)
• 4. estimating the degree of debt tolerance
• (history? Reinhart, Rogoff Savastano,
  NBER 2003)

17

• Debt sustainability
• (Banca dItalia, 2000, 2004, 2006)

• intuition is clear solvency ? debt repayment //
  but timing?
• limt?? dt 0
• moreover, theory says otherwise no Ponzi-game
• limt?? dt (1?)/(1?)-t 0 ? limt?? dt
  ? dlt(?-?)
• (Blanchard et al. OECD90 McCallum, JPE84)
• TltY still leaves many options!
• (Barro JEP89 Kremers JME89, Domar AER44)

18

• Measuring the effects of the cycle

• cyclically adjusted balance (cab)
• cab b - e ?
• ? (y-y)/y
• ? ?b/(?y/y) (?R R/Y - ?G G/Y b) ? G/Y
  se ?R ? 1 e ?G ? 0
• 
  (Bouthevillain et al., ECB 2001 EU avg. 0.9 and
  -0.2)
• problems
• different estimation methods return different
  values of ?
• estimates of ? subject to significant revisions
  (any method)
• estimation of elasticities far from
  straightforward
• is the output gap enough? (composition of
  output other variables)

19

• Dispersion in OG estimates

• large in levels, less so in changes (Orphanides
  e Van Norden, RES 2002)

20

• OG estimates Dispersion and Revisions

21
Why the revisions?

• Y (hence CAB) f (Y past future) therefore
• - revisions of forecasts influence the
  assessment of past outturns
• - the effect can be large in the proximity of
  turning points
• Example 2001 fiscal balances in France and
  Germany

22

• Elasticities

• data intensive
• institutional knowledge vs. econometrics
• (reforms)
• identification of macroeconomic proxies for tax
  bases
• (e.g. profits)

23

• Output composition

• e.g. exports and domestic demand have different
  tax implications
• (Momigliano Staderini, BI98
  Bouthevillain et al., ECB01 Marino et al.,
  QEF08)

24

• A pragmatic approach

• In sum theory does not provide full guidance in
  defining both the long-term anchor and the medium
  term rule for a MTFF
• Need a pragmatic approach
• Define prudent debt levels
• somewhat ad hoc (UK-EU) but not too different
  from proposals by theorists (Blanchard ES90,
  Buiter EP85)
• Derive corresponding structural deficit targets
• - based on long-term expenditure projections
• - use sensitivity analysis
• (sustainability reports Norway EPC ageing
  working group 2006)
• Define medium-term / over-the-cycle targets
  with escape clauses in the face of unfavorable
  circumstances
• (Sweden UK code of fiscal conduct )

25
A few examples
26

• The design of fiscal rules

• Rules commitment-devices and signaling tools
• (they increase the costs of deviating from
  the target for policymakers and
• reduce publics uncertainty about
  policymakers commitment)
• Constrain the bias but mindful not to
• introduce excess rigidity and prevent adequate
  responses
• (e.g. let automatic stabilizers play in bad
  times balanced budget rule?),
• force inadequate responses
• (e.g. a fiscal contraction in response to a
  temporary spike in interest rate, or depreciation
  of the exchange rate),
• Let the bias unchecked in specific circumstances
• (e.g. allow for procyclical expansions
  medium-term/over-the-cycle formulations like
  the (old?) SGP and UK code of fiscal conduct)

27
Other issues of design

• some rules are not targeted to fiscal
  discipline
• (golden rule ? sustainability)
• window dressing
• (inconsistent deficit/debt indicators SGP)
• some rules cannot stand alone
• (e.g. expenditure rules)

28

• Expenditure rules

• PROS
• provide a stronger link between the long-term
  anchor and
• the multi-annual budget exercise
• tackle the bias at source (expenditure)
• let (revenue) automatic stabilizers play
• BUT
• cannot leave the revenue side unchecked
• (tax expenditure)
• MORE COMPLEX THAN IT SEEMS
• what about automatic stabilizers on the
  expenditure side?
• possible exploitation of planning margins?

29
Tax expenditures - Sweden
30
Exploitation of planning margins - Sweden
31
III. Some Remarks about Italy
32
The long-term anchor and the fiscal stance

• SGP debt-to-GDP ratio 60 but when?
• (satisfactory pace never defined)
• Often D/Ylt100 targeted in official documents
  BUT on what basis?
• No sustainability report setting medium-term
  deficit target?
• (long-term projections only for the EPC AWG)

33
The medium-term rule

• EU structural adjustment by ½ percent per year
  towards structural balance free play of
  automatic stabilizers
• Truly endorsed?
• No explicit convention about headroom within the
  target

34
As a consequence. (A)
General Government debt and deficit ( of GDP)
35
As a consequence. (B)
Fonte Banca dItalia, Relazione Annuale 2007
36
Multi-year budget, top-down preparation, tight
execution

• No true multi-year budget
• (3 years but t1 and t2 are forecasts, not
  binding plans)
• No top-down budgeting
• (no expenditure ceilings)
• Rather lax execution
• (no explicit contingency reserve but weak link
  between authorizations from the state budget and
  accounts relevant to the fiscal targets)
• (possibility to use windfall revenues to
  increase expenditure)

37
As a consequence.
Contributions to deficit reduction ( PIL)

• Deficit reduction mostly from higher revenues and
  lower interest exp.

38
Prudent forecasts - Accountability

• Insufficient information on fiscal projections on
  a current programs basis on costing of new
  legislation
• Difficult to assess outturn
• No formal assignment of independent assessment
• (plus nothing much happens if targets are
  not met)
• Budget and legislation by line-items not programs
• Not surprisingly the 2007/2008 spending review
  found abundant evidence of inefficiency in the
  use of public money
• (CTSP08)

39
A few examples
Ministry of Infrastructures Regional Offices
staff per billion of assets (CTSP08)
40
A few examples
Ministry of Justice Prisons Expenditure per
inmate vs. number of inmates (CTSP08)
41
A few examples
Ministry of Justice Courts Elasticity of
scale (yertical) vs. Number of Judges
(horizontal) (CTSP08)
42
Education resources vs. outcomes
An inverse correlation between resources and
outcomes (Montanaro, QEF08)
Proficiency levels (primary and lower secondary
schools (INValSI)
Teachers per pupils
Darker colors higher values
43
Health Services Resources vs. Activities
Another inverse correlation (Francese
Romanelli, BI09)
Expenditure per resident (adjusted for the
composition of population)
Patients mobility
Darker colors higher values
44
IV. Summary and Conclusions
45

• 1. There are significant incentives to fiscal
  indiscipline
• (both theory and evidence)
• 2. This entails both macro-risks and micro
  inefficiency
• 3. MTFFs can help re-engineering incentives and
  control risks
• they are not a magic wand but one is better-off
  having them
• (issues in design enforcement)
• Italy EU fiscal rules provide a frame but
  content needs to be
• defined at the national level

46
1st on the to-do-list Improve Accountability

• Educate the public set credible objectives,
  openly discussed and clearly communicated to
  population at large (e.g. sustainability reports)
• Ex post reconciliation of changes from one budget
  to the next
• Parliamentary scrutiny of financial performance
• Range of sanctions for persistent overspenders

47
What Others Do

• EX ANTE
• External scrutiny of economic assumptions (UK)
• Use of consensus economic forecast (Canada)
• Independent evaluation of macroeconomic
  conditions and fiscal stance (Sweden)
• Independent fiscal forecasts (Netherlands)

• EX POST
• Transparent reporting of fiscal performance (UK,
  NZ)
• Independent evaluation of fiscal compliance with
  objectives (Sweden)
• Predetermined mechanisms for addressing
  deviations form forecasts (Switzerland)

48
Medium-term Expenditure FrameworksRange of
Advanced Country Models
COUNTRY COVERAGE COVERAGE COVERAGE COVERAGE LEVEL OF DETAIL TIME HORIZON DISCIPLINE DISCIPLINE
COUNTRY Soc Sec Debt Interest Local Govt of public spending LEVEL OF DETAIL TIME HORIZON Rolling or Fixed Frequency of Revision
AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS AGGREGATE EXPENDITURE CEILINGS
Finland Some No No 36 Total Spending 4 4 fixed Every 4 years
Netherlands Yes No Tfers 80 4 Sectors 4 4 fixed Every 4 years
Sweden Yes No Tfers 64 Total Spending 3 2 fixed 1 rolling Every year
FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS FIXED MINISTERIAL PLANS
United Kingdom No No Tfers 59 25 Depts 3 2 fixed 1 rolling Every 3 years
France No Yes No 31 35 Missions 3 2 fixed 1 rolling Every 2 years
ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES ROLLING PROGRAM ESTIMATES
Australia Yes Yes Yes 100 20 Depts 267 Progs 3 Rolling Every year
48
49
Effective Multi-year Expenditure
Prioritization Mechanisms
CHARACTERISTIC DEFINITION BEST PRACTICE
Centralized No competing source of expenditure authority Australia
Strategic Assumes policies, laws and contracts can be changed Netherlands
Legitimate Combines bottom-up input from ministries with top-down engagement from politicians France
Comprehensive Covers all expenditure over the whole planning horizon United Kingdom
Capped Operates withing a fixed, multi-year budget constraint Sweden
Definitive The final verdict from the top of the office Finland
50
Adjustment MechanismsManaging risks,
pressures shocks in a multi-year system

• 1. Exclusion
• Excluding volatile/non-discretionary items from
  the rule celing, such as
• debt interest
• unemployment benefits
• social security
• earmarked revenues
• local government (own resources)

• 2. Adjustment
• Adjusting ceilings to accommodate real economy
  effects, such as
• inflation (Netherlands)
• revenue windfalls (Netherlands, Canada)

• 4. Budget Architecture
• Capacity to absorb shocks
• max 20-30 main budget headings
• each budget a mixture of discretionary and
  non-discretionary items
• maximum flexibility to reallocate
• ministerial contingency reserves
• mandatory savings targets

• 3. Contingency Reserves
• Building contingency margins into expenditure
  projections or ceilings
• Netherlands 0.25
• UK 0.75 1
• Canada 1.5 2
• Sweden 1.5 2
• Australia 1.5 5

51
Once again on public scrutiny

• good finance cannot be attained without
  intelligent care on the part of the citizens ...
  due equilibrium between income and outlay will
  only be found where responsibility is enforced by
  the public opinion of an active and enlightened
  community
• Charles Bastable (1927)