Fiscal discipline and infrastructure spending

Transcription

1 Fiscal discipline and infrastructure spending Luis Servén The World Bank Lima, July 2008

2 Fiscal discipline and infrastructure 1. The facts 2. Fiscal discipline and spending composition 3. Rethinking fiscal targets and rules 4. Concluding thoughts

3 1. The facts 2. Fiscal discipline and spending composition 3. Rethinking fiscal targets and rules 4. Concluding thoughts

4 The facts Since the 1990s, drive towards fiscal discipline in LAC to restore macro stability and prevent crises. Generalized increase in primary fiscal surpluses. But in most countries it involved a sharp decline in public infrastructure investment Not only a Latin American phenomenon

5 The facts Latin America

6 The facts 2.00% Brazil Primary fiscal deficit and public infrastructure investment (% GDP) 3.50% 1.00% 0.00% -1.00% -2.00% -3.00% -4.00% -5.00% Primary deficit (left scale) Infrastructure investment (right scale) % 2.50% 2.00% 1.50% 1.00% 0.50% Source: Afonso (2005)

7 2.0 The facts Turkey Primary Deficit and Public Infrastructure Investment (% of GDP) Primary Fiscal Deficit (left scale) Public Infrastructure Investment (right scale) Source: Article IV: Turkey, various years, IMF. Calderón, C., R. Odawara and L. Servén, The World Bank

8 The facts The same trends arise in rich countries (e.g., adjustment to the Maastricht targets) But a bigger concern in emerging and poor countries because they have to build up their infrastructure capital stocks.

9 The facts Public investment contraction is not a concern for growth if the private sector steps in. Private initiative did surge in the 1990s in LAC but: o o o Uneven response across countries and sectors largest in Chile; elsewhere, mostly in telecom. Boom, then bust after 1998 not the big success some hoped. Evidence of public-private complementarity, not only substitution: (Chile, Colombia) Overall, the private response was insufficient for infrastructure to keep up with other developing regions (e.g., East Asia) a widening gap.

10 The facts Brazil: the power sector Investment and asset accumulation Investment Capacity change

11 The facts The widening infrastructure gap Overall Infrastructure quality, regional medians LAC (11) EAP7 (7) MIDDLE (27) IND (21) Question: The quality of the infrastructure is among the best in the world (1=strongly disagree; 7 =strongly agree). Source: Global Competitiveness Report. Source: Calderón and Servén (2008)

12 1. The facts 2. Fiscal discipline and infrastructure 3. Rethinking fiscal targets and rules 4. Concluding thoughts

13 Fiscal discipline and infrastructure Fiscal discipline, as measured and enforced by markets and IFIs, is biased against expenditures accruing future returns such as infrastructure Discipline is based on fiscal rules targeting liquidity -- the overall deficit rather than solvency and net worth, the key to fiscal sustainability. Infrastructure projects typically exert a negative impact on cash flows the cost is immediate but the returns take time to arrive. Stress on public liabilities (gross public debt) and disregard for assets (i.e., future returns) acquired by productive spending today

14 Fiscal discipline and infrastructure Liquidity / cash flow targets distort the timing of revenues and expenditures cut spending today even if that means higher spending tomorrow (e.g., to rebuild the roads). Liquidity / cash flow targets distort the composition of spending they make no distinction between spending with and without future returns. This adds to well-known political economy factors that bias fiscal adjustment against public investment because of the relatively low political cost of investment cuts.

15 Fiscal discipline and infrastructure Fiscal adjustment through infrastructure cuts is like walking up the down escalator (Easterly et al 2007) The direct, short-run effect is to raise cash flows But the indirect effect is to restrict the supply of productive services, growth and hence future tax bases Over time, this may undo much of the fiscal adjustment and require further cuts, even less growth and so on This is especially important in countries where poor infrastructure is a major obstacle to growth

16 Fiscal discipline and infrastructure The fiscal impact of public infrastructure: three ingredients at play the marginal productivity of infrastructure depends on the government s ability to select good projects and the (relative) scarcity of infrastructure services; the user cost of public infrastructure capital depends on the efficiency of public procurement (e.g., waste and corruption), the marginal cost of borrowing, and the maintenance cost; the government s ability to capture the returns depends on both cost recovery via user fees and tax collection capacity These ingredients vary a lot across countries (e.g., Chile vs Nicaragua) and types of projects (e.g., utilities vs roads).

17 1. The facts 2. Fiscal discipline and infrastructure 3. Rethinking fiscal targets and rules 4. Concluding thoughts

18 Rethinking fiscal targets and rules Private initiative offers much potential but the public sector will remain a key player in infrastructure for years to come (except perhaps in telecom) There is a need to reduce the bias against productive public spending. Two basic options: Option 1: Make exceptions to existing targets and rules Give special treatment to some public investment projects, exempting them from cash-flow targets. Option 2: Develop new fiscal indicators

19 Rethinking fiscal targets and rules Option 1: Keep the targets but make exceptions What exceptions? (a) Exclude commercially-oriented public enterprises from fiscal targets Not a good idea if those firms have positive cash flows the rest of the public sector will face harder targets What s commercially run? Hard to specify objective criteria in practice. Few if any such enterprises found in Latin America anyway (IMF 2005), so largely a non-issue.

20 Rethinking fiscal targets and rules Option 1: Keep the targets but make exceptions (b) Exclude projects with private sector financing Potential for efficiency gains -- but PPPs often used to place projects off-budget rather than for efficiency reasons. They create the appearance of fiscal discipline by hiding the fiscal costs (guarantees, long-term commitments) No established accounting standards for the fiscal repercussions -- that have been big in some episodes. The result is a weakening of fiscal transparency and accountability. PPPs much trickier than thought -- strong technical and institutional demands on the public sector.

21 Rethinking fiscal targets and rules Option 1: Keep the targets but make exceptions The exceptions approach might help but 2 major problems: Incentive to distort (or misclassify) spending to make it fit under the exceptional categories Investment program biased towards the permitted categories not necessarily the priority items (e.g., mining vs roads).

22 Rethinking fiscal targets and rules Option 2: Develop new fiscal indicators Traditional fiscal indicators are useful Short-term cash deficit indicates liquidity Debt indicates vulnerability to shocks but they don t tell the whole story Don t indicate if cash is being used to pay for current consumption or assets that deliver future benefits Don t offer information on government s stock of assets A more complete set of indicators would include measures of net worth and its changes How can they be constructed?

23 Rethinking fiscal targets and rules Option 2: Develop new fiscal indicators Accrual accounting allows measuring net worth (balance sheet) and its changes (income statement) Not expensing investment in full, just asset depreciation like in corporate practice. Established standards: International Public Sector Accounting Standards, based on IFRS; IMF s Government Finance Statistics Manual 2001 Many countries shifting to accruals basis. But there are limitations: Accounting asset values don t always match true values; assets may generate no future cash flows Future taxes and social security payments excluded

24 Rethinking fiscal targets and rules Option 2: Develop new fiscal indicators Long-term cash flow projections (e.g. 30- or 50-year) can help with some of these problems Can include taxes and future social security payments and effects of public investment on future tax revenue But the government may be able to manipulate the projections to generate the numbers it wants how can this be avoided? Transparency: make assumptions and models public and allow analysts to produce their own estimates Develop standards for the projections and a system for auditing them Outsource projections to an expert council (like with Chile s structural surplus)

25 Rethinking fiscal targets and rules Benefits and drawbacks of three sources of fiscal indicators Cash accounting Modern accrual accounting Long term projections Provides information on short term cash flows Provides information on net worth, given current policies Yes Yes Yes No Partially Yes Incorporates uncertainty Avoids the issue Partially Yes Limits self serving forecast bias Yes Yes Not easily Source: Easterly, Irwin and Servén (2008)

26 Rethinking fiscal targets and rules Option 2: Develop new fiscal indicators Accrual accounting allows implementing the golden rule a current balance target: permits debt (up to a ceiling, as in the U.K.) to finance capital (but not current) spending. Does not guarantee solvency unless the investments are financially sound and does not help with O&M spending May encourage misclassification of current expenditure as investment A more conservative golden rule (Mintz and Smart 2007): Only cash-generating assets should be treated as investment Government must fund some proportion of the cash-generating assets from taxes (i.e., less than 100% debt finance)

27 Rethinking fiscal targets and rules Option 2: Develop new fiscal indicators Long-term projections allow targeting net worth itself -- the permanent balance rule, in analogy with the permanent income hypothesis in consumption theory (Buiter) Comprises assets and liabilities so no spending distortion. But calculations of future income and expenditure are complex and open to cheating. Challenging technical and institutional requirements Implementing explicit net worth targets is not within easy reach for most countries (only New Zealand, Norway so far)

28 Rethinking fiscal targets and rules But room for more infrastructure spending ( fiscal space ) will not help much without better spending. Infrastructure projects are especially prone to waste, corruption, rentseeking, clientelism. Political economy favors too much new investment relative to O&M (photo ops) Other things equal, countries with weaker governance show higher public investment (Keefer and Knack 2007). Many white elephants acquired in the name of public investment There is a need for sound project selection and evaluation so that only high-return projects are undertaken still a pending challenge in most LAC countries (except Chile) Financial returns need to be captured through user fees and/or taxation still too low in most countries.

29 1. The facts 2. Fiscal discipline and spending composition 3. Rethinking fiscal targets and rules 4. Concluding thoughts

30 Concluding thoughts The public sector will continue to play a key role in infrastructure for years to come so the anti-investment bias of fiscal programs is a concern. Fiscal programs tend to focus too much on liquidity and too little on solvency, often leading to excessive public investment compression and future growth losses. Growth reductions are as much of a fiscal risk as deficit increases, so fiscal programs need to reconcile better solvency and growth -- paying more attention to the composition of spending and the measurement of net worth.

31 Concluding thoughts Liquidity and cash-flow do matter too but they are distinct from solvency and cannot be measured by the same statistic (the conventional overall balance). Accrual accounting and long-term projections can be useful to track solvency and net worth, and help improve the design of fiscal programs but they require high technical capacity and strong fiscal institutions. However, there is increasing recognition that the technical and institutional demands of PPPs are no less challenging. More spending will be of little help without institutional reforms to ensure better spending

32 End

33 Rethinking fiscal rules Benefits and drawbacks of three sources of fiscal indicators Cash accounting Modern accrual accounting Long term projections Provides information on short term cash flows Provides information on net worth, given current policies Yes Yes Yes No Partially Yes Incorporates uncertainty Avoids the issue Partially Yes Limits self serving forecast bias Yes Yes Not easily Source: Easterly, Irwin and Servén (2008)

34 The facts 4% European Union: primary deficit and public investment (percent of GDP, average of 11 Maastricht countries) 4.0% Primary Deficit 2% 0% -2% -4% -6% Primary Deficit Public Investment 3.5% 3.0% 2.5% 2.0% Public Investment No te: the co untries are A ustria, B elgium, Finland, France, Germany, Italy, Netherlands, P ortugal, Spain, Sweden, and U.K. Source: OECD Economic Outlook database.

35 The facts Sub-Saharan Africa 5 Infrastructure Expenditure and Overall Deficit (% of GDP, 11 countries, GFS data) Infrastructure Deficit Source: Estache (2005)

36 The facts 80,000 Latin America: Investment in Infrastructure Projects Investment in LAC Infrastructure Projects with Private Participation with Private Participation (2000 US$ million) (commitments, 2000 US$ million) 70,000 60,000 50,000 40,000 30,000 20,000 10, Source: World Bank PPI database, 2008

37 The facts Infrastructure investment in Latin America 5% (6 major economies, GDP-weighted average) 4% Total Public Private 3% 2% 1% 0% Source: Calderón and Servén (2008)

38 3.0% Infrastructure investment in Latin America (6 major economies, GDP-weighted average) Power Land transport 1.5% 2.5% 2.0% 1.0% 1.5% 1.0% 0.5% 0.5% 0.0% % Total Public Private 1.5% Telecommunications 1.5% Water Total Public Private 1.0% 1.0% 0.5% 0.5% 0.0% % Source: Calderón and Servén (2008)

39 The facts 8% Infrastructure investment in Latin America (6 major economies, GDP-weighted average) 7% 6% 5% 4% 3% 2% 1% 0% Chile Colombia Source: Calderón and Servén (2008)

40 The facts Infrastructure investment in Latin America 6% 5% (6 major economies, GDP-weighted average) Argentina Brazil Mexico Peru 4% 3% 2% 1% 0% Source: Calderón and Servén (2008)