Fiscal Policy and Long-Term Growth – Case Studies

This paper explores how fiscal policy can affect medium- to long-term growth.

Abstract

This paper explores how fiscal policy can affect medium- to long-term growth.

Introduction

1. This supplement presents the case studies on the impact of fiscal reforms on long-term growth. The sample of countries comprises Australia, Chile, Germany, Ireland, Malaysia, Netherlands, Poland, Tanzania and Uganda. Annex I outlines how both quantitative and qualitative indicators were used to select countries on the basis of the number of reform episodes, income groups and geographical diversity. The studies cover a wide variety of fiscal reforms, including macroeconomic stabilization, tax policy, expenditure policy and institutional reforms. With regard to transmission channels to growth, they consider three channels: human capital-adjusted labor supply, physical capital, and total factor productivity. The lessons from the case studies are integrated into the main Board paper discussion.

2. The studies apply the Synthetic Control Method (SCM) to assess the effect of fiscal policy reforms on GDP growth. The SCM is a data-driven technique to quantify the impact of an event (fiscal reforms in this case) on an outcome variable of interest (long-term growth in this case). With the SCM, a counterfactual, or “synthetic” country, is constructed as a weighted average of a large number of countries that have similar characteristics to the country of interest, but are unaffected by the fiscal policy reform. Once the counterfactual is created, the post-reform growth performance in the country of interest is compared to the growth performance in the counterfactual and the estimated impact of the fiscal reform is then the difference between the two series.

3. The limitations of the SCM method suggest interpreting the results with caution. It is difficult to disentangle the growth impact of reforms from that of various other factors affecting growth. Although econometric tests are used to check the robustness of the results (Annex II), the difference between post-reform growth in the country of interest and its synthetic counterpart could be due to factors other than fiscal policy. Results could also be potentially biased if the comparator groups include countries that also underwent growth-enhancing reforms. In this case, any potential bias introduced by including countries that also undertook reforms should reduce the impact of the fiscal reform on growth meaning, in this regard, the estimates are conservative.

Limitations of the SCM notwithstanding, country studies suggest that fiscal reforms can affect long term growth.

  • Impact. In advanced economies (excluding Ireland), long-term per-capita growth is about ¾ percentage points higher than the counterfactual. In emerging markets and low-income countries long-term growth is 2¾ and 1¾ percentage points higher, respectively (Figure 1a).

  • Transmission channels: The relative contribution of each growth channel varies by income group. For emerging market and low income countries capital accumulation was the most significant driver while the advanced economies studied had relatively balanced contributions from each channel (Figure 1b).

Figure 1.
Figure 1.

Evidence from Country Studies: Long-Term Growth Post-Reform

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source; IMF staff calculations; Supplement 1.1/ 5-year averages for Germany and Poland.2/ Chile (1) refers to the first reform episode (1974); Chile (2) to the second reform episode (1983); Australia (1) to the first reform episode (1985); and Australia (2) to the second reform episode (1998).

Lessons:

  • Design matters for the successful implementation of reforms. In all countries studied the design of fiscal measures reinforced each other. For example, in the Netherlands, Ireland, and Germany, work incentives were enhanced by reducing benefits and income taxes. Fiscal reforms were also commonly implemented together with broader structural reforms to maximize their growth impact.

  • Social dialogue enhances the likelihood of reforms being implemented and sustained. Various strategies were used to foster public support including effective communication with stakeholders (Germany) and compensatory measures for those expected to lose from reforms (Ireland, Netherlands). Common policy anchors, such as the prospect of EU accession in Poland, helped forge consensus.

Australia: Fiscal Policy and Long-Term Growth

A. Background

1. Australia established a strong track record of steady economic reforms over several decades. This process of continuous economic reform began in the mid-1980s, following a decade of comparatively poor growth; high unemployment; persistent inflation; deteriorating competitiveness and rising external and fiscal imbalances. Early efforts aimed at opening up the economy through tariff reform; and exchange rate and financial sector liberalization (Debelle and Plumb, 2006). However, these reforms were insufficient to establish a sustained improvement in growth. These initiatives were followed by a successful fiscal consolidation, which began in 1985, and which was built upon comprehensive tax reform and credible fiscal rules. In the 1990s, the economic strategy turned to the real sector. The government strengthened product market competition and liberalized wage bargaining arrangements. In 2000, the focus of reform efforts returned to fiscal policy. A new consumption tax, the Goods and Services Tax (GST), was introduced, allowing limited reductions in personal income tax rates. The reform also permitted the abolition of the inefficient wholesale sales tax and provided a more reliable fiscal equalization arrangement for Australian States.

2. In many areas, such as tax policy, pensions, and fiscal transparency, Australia was an “early adopter” at the forefront of designing and implementing innovative reforms. Wide-ranging fiscal reforms were conducted in tandem with major product and labor market reforms. In the mid-1980s it adopted an ambitious tax reform that reduced personal income tax rates. In 1992, it overhauled its pension system when a fully funded three pillar superannuation scheme replaced the old “pay-as-you-go-system.” In 1998, a Budget Honesty Act was passed, introducing transparent budget reporting, intergenerational accounting, and medium-term fiscal planning. These fiscal reforms took place within the context of a wider reform effort that also included substantial improvements of wage bargaining arrangements, strengthening product market competition and enhancing intra-governmental fiscal relations.

3. Australia’s federal structure, comprising of the national government, six states and two territories, also strongly influenced the design of economic reforms. Ensuring the successful implementation of reforms required a high degree of consensus-building and strong intra-governmental cooperation. In tandem with economic reforms, the country developed institutions that provide opportunities for discussing policy challenges and coordinating the national reform agenda across different levels of government.

4. Australia’s economic reforms took place against a backdrop of a strongly improving external environment. The rapid increase in foreign demand for Australian commodities, along with falling import prices, led to a substantial improvement in the terms of trade. The commodities boom, particularly over the last decade, fueled a rapid increase in mining-related investment, which bolstered economic growth.

5. Nevertheless, growth decelerated somewhat over the last decade relative to 1990s; renewing the focus on economic reform (Figure 1). This deceleration provoked a reappraisal of the policy framework and a search for new measures that will reinvigorate growth. The future of fiscal policy figured prominently in this reappraisal. In 2010, the Australian Government commissioned a thorough review of tax policy, as well as an examination of federal fiscal relations. One major early result of this review has been a fundamental rethinking of the role of natural resource taxation.

Figure 1.
Figure 1.

Australia: Selected Indicators, 1970–2011

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Australian authorities, IMF staff calculations.

Fiscal Policy Innovation and Commitment to Reform

A. Overview

6. In the 1970s and early 1980s, Australia suffered from severe macroeconomic imbalances. Growth was low and volatile relative to other advanced economies, due to the economies’ dependence on commodity exports. Double-digit inflation was a persistent problem, while unemployment increased from less than 2 percent in 1971 to almost 10 percent in 1983.

7. These poor economic outcomes led to widespread dissatisfaction with the development model (Fenna, 2013). Australia was a comparatively closed economy, with a smaller share of its GDP being traded than would be normal for an economy of its size (Anderson, 2009). The terms of trade deteriorated throughout the 1970s as the prices of commodity exports declined, adding downward pressure on the growth rate. Many sectors of the economy, particularly manufacturing and agriculture, were heavily regulated or benefited from protectionist policies. By the end of the decade, a combination of global shocks, a deep recession, and growing fiscal and external sector imbalances emphasized the need for fundamental structural adjustment. As such, early fiscal reform efforts focused on tariff reforms in order to strengthen external competitiveness.

8. The new government elected in 1983 launched a comprehensive economic strategy that aimed to reestablish macroeconomic stability and reinvigorate economic growth. This strategy had three broad objectives:

  • Improving the tools of macroeconomic management. The Australian dollar was allowed to float; capital controls were removed; and interest rates were liberalized.

  • Restoring macroeconomic stability. Continued large fiscal deficits prompted renewed efforts at fiscal consolidation in 1985. To add credibility to their deficit reduction efforts, the Hawke-Keating government adopted a ‘trilogy’ of fiscal rules which became operational in the 1986 budget. The government made a commitment to not increase either taxes or expenditures as a ratio of GDP, and to reduce the budget deficit both in absolute terms and as a ratio of GDP.

  • Strengthening the supply side of the economy. A four year tax reform package was also introduced that established new taxes on capital gains and fringe benefits. The wholesale sales tax was streamlined while tax expenditures on sectors such as agriculture, forestry and film were reduced. The corporate tax rate was lowered progressively from the late 1980s onwards (Figure 6). To encourage greater labor force participation, income tax rates were reduced. Between 1987 and 1992, the tax burden on individuals fell from 13.5 percent of GDP to 11.4 percent (Table 3). To reorient the economy to external markets tariff reform was accelerated.

Figure 2.
Figure 2.

Australia: GDP Growth, 1970–2014

(Percent, annual)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: WEO.Note: shaded blue area indicates reform period; orange area indicates global financial crisis.
Figure 3.
Figure 3.

Australia versus European Union-15 +3

(Annual GDP growth, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ European Union-15 (excluding Ireland, Netherlands and Portugal), Canada, United States and Japan.
Figure 4.
Figure 4.

Synthetic Control Method: Australia

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.Note: Shaded blue areas indicate reform periods.1/ United States, Spain, Greece and United Kingdom.2/ Sweden, United States, Spain, Greece and United Kingdom.3/ United States, Sweden, Ireland, Greece and Japan.4/ United States, Sweden and Greece.
Figure 5.
Figure 5.

Australia: Growth Decomposition, 1973–2007

(Percent, 3-year moving average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Penn World Table.Note: shaded blue areas indicate reform periods.
Figure 6.
Figure 6.

Australia, Fiscal Policy and Growth, 1970–2012

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Australian authorities, Historical Debt Database, and IMF staff calculations.Note: Shaded blue areas indicate reform periods; orange areas indicate global financial crisis.
Table 1.

Australia: Major Fiscal Reforms

article image
Source: IMF staff reports.
Table 2.

Australia : Economic Growth Predictor Means before Fiscal Reform

article image
Source: IMF Staff calculations.

All variables except GDP growth and GDP growth per capita are averaged for the 1970–1984 period. The average for GDP growth and GDP growth per capita is calculated using the years 1970, 1977 and 1984.

All variables except GDP growth and GDP growth per capita are averaged for the 1970–1997 period. The average for GDP growth and GDP growth per capita is calculated using the years 1970, 1983, 1997.

Table 3.

Australia: Selected Indicators

(Period Averages)

article image
Source: Australian authorities, Historical Public Debt Database, Penn World Table, SWIID 5.0, WEO.

Growth accounting estimates are calculated using Penn World Table 8.0.

“Market” Gini measures the income distribution before taxes/transfers; “net” refers to after taxes/transfers.

9. Australia was one of the first advanced economies to tackle demographic change with fundamental pension reform. In 1992, the Keating Labor government introduced a three pillar pension reform called “Superannuation Guarantee”. The reform included a means-tested “safety net” pension, the establishment of a compulsory employer-based contribution that funded private pension schemes for all employees, and a voluntary pension pillar. Since the pension system was introduced, Australia’s pension investment funds have grown dramatically. In 1992 assets totaled AUS$150 billion. By June 2014, total superannuation assets amounted to $1.85 trillion (ASFA, 2014). The reform is associated with a substantial increase in private sector investment to GDP ratio.

10. This first phase of fiscal reforms ended with a severe recession in 1991, leading to another period of deteriorating fiscal deficits (Table 2). This recession coincided with a global downturn, but it was exacerbated by a collapse in Australian asset prices, including housing, leading to large losses in the banking sector (Gizycki and Lowe, 2000). A further round of fiscal consolidation resulted in a return to budget balance by the mid-1990s. Efforts were made to contain public indebtedness, which was in part financed through privatization (Gruen and Sayegh, 2005). Thereafter, the government recorded surpluses for the following eight years.

11. The 1991 recession prompted another round of economic reforms that focused on the real economy (Table 1). Inflation targeting was introduced in 1993. In 1995, there was an overhaul of the legislation governing product market competition. A coordinated National Competition Policy was agreed by all governments in Australia’s federal system, extending anti-competitive conduct laws to cover previously exempt government and unincorporated enterprises. Barriers to entry in highly regulated sectors like telecommunications were reduced while the regulatory framework governing monopolies was radically changed (Kerf and Geradin, 2005).

12. The 1990s also witnessed a gradual but comprehensive restructuring of the wage bargaining system (Wright and Lansbury, 2014). The Workplace Relations Act 1996 reduced the role of the Industrial Relations Commission in setting wages and conditions—primarily by restricting its role to adjudicating on “safety net” wage adjustments. Labor market policies also encouraged the decentralized wage bargaining (Dawkins, 2000). In 1990, only a third of Australian workers had wages set through enterprise-based or individual contracts. By 2002, this ratio had increased to almost 80 percent (Productivity Commission, 2005). These labor market reforms paid a crucial role in curtailing real wage growth, thus re-establishing external sector competitiveness.

13. The late 1990s saw a second round of fiscal policy innovations, which focused on fiscal transparency and tax reform. The Charter of Budget Honesty, launched in 1998, established a new framework for the conduct of fiscal policy (Miyazaki, 2014). The Charter also improved transparency by facilitating public scrutiny of fiscal outcomes. Governments were obliged to provide regular updates of fiscal out-turns, including a mid-year Economic and Fiscal Outlook. A final budget report had to be published within three months after the end of the financial year. As soon as an election was called, the Charter required the Departments of Treasury and Finance to prepare an independent report—the Pre-Election Economic and Fiscal Outlook. Governments were also obliged to specify their long-term fiscal objectives within which short-term fiscal policy would have to operate. To facilitate a broader understanding of demographic related fiscal challenges, the act obliged the government to publish an Intergenerational Report (IGR) at least once every five years.

14. In July 2000 a broadly-based consumption tax, known as the Goods and Services Tax, was introduced. It replaced the previous Federal wholesale sales tax system as well as a large number of heavily distortionary State and Territory Government taxes, duties and levies such as banking taxes and some stamp duties. The higher tax revenues allocated to the States also allowed the federal government to remove government financial assistance grants allocated to the states.1 The additional revenues from the GST also facilitated a shift of the tax burden away from direct taxation, both for personal and business taxpayers. Average tax rates were reduced for all income levels in 2001 (McDonald and Kippen, 2005).

15. More recently, tax reform has again emerged as an important issue on the policy agenda. In 2008, the government initiated a wide-ranging review of the tax system in order to identify a reform agenda for the next twenty years. This review led to the proposal to introduce an ambitious Resources Super Profit Tax. Subsequently, a controversial Minerals Resource Rent Tax was established but was repealed in 2014.

B. Notable Design Features

16. Policies promoting social cohesiveness played a decisive role in the success of Australia’s reform strategy. The highly decentralized political structure means that sub-national entities are important actors in the policy dialogue. This necessitated the development of a “Collaborative Federalism” model, where formal institutions have been created to enhance cooperation between different levels of government.

17. A key step in this process was the creation of the Council of Australian Governments (COAG) in 1992. This body replaced the Premiers Conferences where State Prime Ministers met regularly to discuss matters of common interest. Since its creation, the Council has taken a leading role in developing a national reform agenda over wide areas of common interests. For example, in 2009, the COAG played a leading role in developing a new Federal Financial Relations Framework.

18. The introduction of the consumption tax in 2000 provided a good example of successful intra-governmental cooperation. The new tax was accompanied by a parallel reform of the State Government financing model, providing the necessary income that permitted the abolition of a large number of highly distortionary state-based fees and taxes. This allowed the development of a powerful coalition of interests across different levels of government in support of an important policy reform.

19. Australia also benefited from strong civil and public institutions that have fostered an environment where policy options can be developed. Think tanks, trade unions, the financial press, universities, as well as institutions such as the National Productivity Commission have all actively contributed to the dialogue over the challenges facing Australia, facilitating a high degree of policy innovation. In this regard, measures to enhance fiscal transparency have helped ensure that the political debate over policy options is more informed.

Impact on Long-Term Growth

20. Australia has enjoyed several decades of strong growth; monetary stability; low inflation; and prudent macroeconomic management (Figure 2). It has one of the highest per capita GDP levels in the world. Since 1980, real GDP growth averaged 3.2 percent per year, while unemployment was consistently below the OECD average. Moreover, Australia’s fiscal outcomes compare favorably to other advanced countries. Since the early 1970s, Australia consistently maintained a general government gross debt below 35 percent of GDP.

21. In order to assess the impact of fiscal policy on economic development, it is useful to consider five periods between 1970 and 2007:

  • The pre-reform period 1970–84. The economy was comparatively inward looking, with a high degree of protection to manufacturing and agricultural sectors. During this period there were very limited fiscal reforms, primarily focused on tariff reduction.

  • The first wave of fiscal reforms 1985–92. The period started with a major consolidation effort, tax cuts, and fiscal rules and concluded with the creation of a fully funded pension system.

  • Real sector structural adjustment 1993–1997. The reform effort was directed towards labor and product markets.

  • The second wave of fiscal reforms 1998–2001. The adoption of greater fiscal transparency and the introduction of the GST, a new funding arrangement for States, and lower personal income tax rates.

  • The post-fiscal reform period 2002–07. This period marks an interregnum between the reforms of the early 2000s, and the last major review of the tax system2 as well as the introduction of a new framework for natural resource taxation.

22. Figure 3 offers a simple comparison between Australia’s per capita growth with a peer group comprising of the EU15 plus major commodity producers—Canada and South Africa— as well as United States.3 During the pre-reform period, a small negative growth differential vis-à-vis the peer group had emerged on average, while Australia suffered greater volatility in growth and rising imbalances, highlighting the motivation behind the structural adjustment that began from 1985 onwards. Economic performance improved in tandem with the first wave of major fiscal and structural reforms. The third period marks an interregnum in fiscal reform efforts. The reform priorities shifted towards the real sector. Growth performance improved during this period, opening up a substantial positive differential with the peer group.

23. Between 1998 and 2001, Australia launched a second wave of reforms, centered on improving fiscal transparency and tax reform. During this period, we see the beginning of a deceleration of growth. The Australian economy slowed compared to the period 1993–97, while its peer group enjoyed a period of accelerating growth. Once this second fiscal reform effort is concluded, the Australian economy continues decelerate. Nevertheless growth performance compares favorably in relative terms to the peer group with a positive growth differential increasing.

24. A counterfactual growth series was generated for a hypothetical Australia (Figure 4 and Table 2). It was constructed using a synthetic control group—comprising of advanced economies that did not experience substantial fiscal reforms in periods similar to Australia’s— and a set of common growth predictors. Given the multifaceted nature of the growth process, the results of the synthetic control method need to be interpreted with care. Nevertheless, the approach offers some interesting insights into the Australian growth experience. The counterfactual confirmed that a growth differential opened up after 1985, when the first fiscal reform wave started. In the mid-1990s the growth differential between Australia and the counterfactual series narrows, and briefly reversed as Australian growth decelerates during the second fiscal reform wave, while the counterfactual growth rate picks up. This second wage of reforms did not establish a higher growth rate relative to the pre-reform period; Australian growth continued to decelerate. However, the reforms did reestablish a small, but positive differential vis-à-vis the peer group during 2002–07.

25. The results of the synthetic control method were tested with three robustness checks. The first test, the placebo test, applies the synthetic control method to every country in the comparator group (Figure A1). Post-reform growth in Australia was higher relative to the growth distribution of non-reformers (countries in the comparator group), indicating that the fiscal reform likely has made a positive difference in the growth performance in Australia. The second test examined the implications for the results by sequentially excluding countries used to estimate the synthetic growth series (Figures A2). This test indicates that the results are robust—post-reform Australian growth was higher than the synthetic growth series in each of the cases where a country is excluded. The third test examined how changes in the pre-reform sample size affected the results (Figures A3, A4, A5, and A6). The baseline assumption is that 1985 marked the beginning of the substantive fiscal reforms in Australia, when the Hawke-Keating fiscal consolidation began with the establishment of the “trilogy of fiscal rules.” However, substantive fiscal reforms also took place in subsequent years, suggesting that an alternative identification of reform periods might be appropriate. This check suggests that the results are robust to alternative starting and ending dates of the pre-reform period. In all cases, the Australian growth out-turn is higher than that suggested by the alternative synthetic control series.

26. A decomposition of growth in terms of capital, labor and total factor productivity (TFP) provides further insights how fiscal policies were transmitted into higher economic growth (Figure 5). The most striking feature of this decomposition is the gradually increasing contribution of capital to economic growth, emphasizing how macroeconomic and fiscal stability improved the investment climate. In particular, pension reform provided a significant increased private sector savings, thus contributing to capital accumulation. In the decade just prior to the first wave of fiscal reforms; capital accumulation contributed on average 30 percent towards GDP growth. During the first decade of this century, this contribution had increased to almost 50 percent. This rising contribution of capital is particularly noticeable after 1992. This was in part due to the strong improvement in terms of trade, as commodity prices increased, prompting a large increase in mining-related investment.

27. The decomposition also suggests that the second wave of fiscal reforms had a somewhat lower impact on economic growth. Between 1987 and 1992, the contribution of labor was the primary growth driver; highlighting the importance of tax changes aimed at reducing economic distortions such as lower personal income tax rates. During this period, labor force participation increased in response to these tax cuts. A more muted, but nonetheless important increase also occurs in the mid-1990s, which was associated with important policy changes that enhanced labor market flexibility. However, these policies did not generate the same impressive improvement in labor force participation experienced in the 1980s. Likewise, we do not see an increase in the labor contribution to growth after the second wave of fiscal reforms from 2000 onwards.

Lessons: Fiscal Reform as a Catalyst for Accelerating Growth

28. Fundamental structural changes, including those in the fiscal area, have ensured that the Australian economy has a greater degree of flexibility than was the case in the 1970s and early 1980s. This has made it more resilient to external shocks such as the recent global financial crisis. Nevertheless, we have also seen a gradual growth deceleration following the first wave of fiscal reforms. This prompted a renewed discussion within Australia regarding the relationship between fiscal policy and growth. Australia recently conducted a thorough review of the tax system, and while many of the measures have not yet been implemented, tax reform continues to be an important policy issue.

29. Fiscal reforms have leveraged all four channels, whereby fiscal policies can influence long run growth:

  • Macroeconomic stability: Two successful fiscal consolidation efforts, coupled with the adoption of fiscal rules have ensured that debt and deficit levels have been contained.

  • Social cohesion: Building political consensus – particularly across different levels of government and civil society – is an essential element explaining the success of many economic policies.

  • Expenditure reforms: The establishment of a fully funded pensions system increased private sector savings and contributed to a sustained recovery in private investment.

  • Tax reform: Lower personal income tax rates helped increase labor force participation rates, positively contributing to growth rates.

30. Developing a fiscal reform agenda in the context of a highly diverse society with decentralized political structure poses particular challenges. Australia created the necessary cooperative institutions—both formal and informal—to ensure a high degree of social cohesiveness and consensus across all levels of government as well as civil society. It established an environment where policy options can be carefully considered, which in turn has allowed Australia to be an innovator in fiscal reform. Moreover, this social discourse has at crucial moments offered a frank and honest assessment of difficulties and challenges facing the economy and; therefore, provided a basis for developing the political consensus around difficult and painful policy measures.

Appendix. Robustness Tests for Australia

A. Placebo Tests

Figure A1.
Figure A1.

Placebo Test: Australia

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.Note: Shaded blue areas indicate reform periods.1/ European Union-15 (excluding Ireland, Netherlands and Portugal), Canada, United States and Japan.2/ European Union-15 (excluding Germany), Canada, United States and Japan.

1. Every country in the comparator group was alternatively chosen as the tested country when in fact no fiscal reform took place in that country (Figure A1). If the estimated effect for Australia is significantly larger than the placebo countries, it strengthens the case that the growth gap estimate in the baseline can indeed be attributed to the fiscal reform. Post-reform growth in Australia was higher relative to the growth distribution of non-reformers (countries in the comparator group), in terms of both the size of growth gaps and ratios of post-reform and pre-reform root mean square prediction errors (RMSPEs). This indicates that the fiscal reform likely has made a positive difference in the growth performance in Australia.

B. Sequential Exclusion of a Comparator Country

2. The sensitivity of the baseline results was tested by sequentially excluding each of the countries in the comparator group that received a positive weight and re-estimating the model (Figure A2 and A3). Post-reform GDP growth gaps are larger than pre-reform GDP growth gaps for almost all cases for both 1985 and 1998 reforms, confirming the robustness of the baseline results.

Figure A2.
Figure A2.

Sequential Exclusion of a Comparator Country: Australia, 1985

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ United States, Sweden and Japan.2/ Sweden, United States, Greece and United Kingdom.3/ United States, Italy, Denmark, Luxembourg and Greece.4/ United States, Sweden and Greece.5/ Canada, Sweden, Greece and United Kingdom.
Figure A3.
Figure A3.

Sequential Exclusion of a Comparator Country: Australia, 1998

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ United States, Sweden, Japan.2/ United States, Sweden, Japan and Ireland.3/ United States, Ireland, Sweden, Italy and Finland.4/ United States, Italy, Ireland, Japan and Spain.5/ Canada, Ireland and Japan.

C. Changes in the Reform Periods

3. Synthetic series were re-estimated for years 1980, 1984, 1986, and 1990 (first wave) and 1993, 1997, 1999, and 2003 (second wave) (Figure A4 and A5). If the growth gap estimate in the baseline changes substantially in response to a slight variation of the starting year, it would cast doubt on the existence and the size of the positive growth gap. If the growth gap does not change substantially in response to a large variation in the starting year, it would undermine our confidence that the positive growth gap is indeed indicative of the effect of the fiscal reform. For the 1985 reform, starting periods are varied from 1980 to 1990, while for the 1998 reform, starting periods are varied from 1993 to 2003. Results show that the conclusion for the baseline is not sensitive to small changes in starting periods, but large changes sometimes result in disappearance of growth effects. These results are consistent with the conjecture in the baseline that the fiscal reform is the likely reason for the positive growth effect.

Figure A4.
Figure A4.

Change in Reform Period-Start Year: Australia, 1985

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ United States, Sweden, Spain and Japan.2/ United States, Sweden and Greece.3/ United States, Sweden and Greece.4/ United States, Spain and Japan.
Figure A5.
Figure A5.

Change in Reform Period- Start Year: Australia, 1998

(Annual GDP growth versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ United States, Sweden, Japan, Ireland and Greece.2/ United States, Japan, Ireland and Greece.3/ United States, Sweden, Japan and Ireland.4/ United States, Sweden, Japan and Ireland.

4. The difference in average growth rates was recalculated by setting the end of the fiscal reform period to different years (Figure A6 and A7). The purpose of this exercise is to examine whether our conclusion in the baseline is sensitive to the timing of the end of the reform, given its uncertainty. Results confirm that the conclusion holds for the 1998 reform when the end of the fiscal reform period changes but not for the 1985 reform.

Figure A6.
Figure A6.

Change in Reform Period-End Year: Australia, 1985

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Sweden, United States, Spain, Greece and United Kingdom.
Figure A7.
Figure A7.

Change in Reform Period-End Year: Australia, 1998

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ United States, Greece and Denmark.

Chile: Fiscal Policy and Long-Term Growth

A. Background

1. The Chilean economy expanded at moderate rates between 1964 and 1970, suffering from stubbornly high inflation and the legacy of protectionism (Figure 1). Real GDP and per-capita GDP grew on average at about 4 and 2 percent, respectively, as several decades of protectionist trade and industrial policies practically isolated the Chilean economy from the world economy. Inflation—a chronic problem of the country—fluctuated around 28 percent per year, on average, reflecting wage increases in excess of productivity gains (Table 1).

Figure 1.
Figure 1.

Chile: Pre-Reform Economic Indicators, 1960–1973

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Banco Central de Chile, DIPRES, WDI, and WEO.
Table 1.

Chile: Selected Indicators, 1964–1973

(Percent change unless otherwise indicated, annual)

article image
Source: IMF staff calculations, WEO.

General government.

Exports of goods and services plus imports of goods and services divided by GDP.

2. During this period, fiscal imbalances were contained thanks to buoyant copper revenues that offset the detrimental impact of interventionist policies. After posting average deficits of about 5 percent per year in the early 1960s, the fiscal balance of the general government improved gradually, benefiting from historically high prices of copper. By 1969, the deficit had shrunk to 0.8 percent of GDP. On the structural front, Chile continued to follow policies centered on import substitution and an active role of the state in the economy. As a result, no significant structural reforms were implemented until 1970.

3. In late 1970, the Allende government adopted aggressive fiscal measures to stimulate demand and redistribute income, but after a rebound the country fell into deep crisis. Fiscal measures included a widespread program of nationalization and a sharp increase in public transfers, wages, and social spending on housing, health, and education. As a result, the deficit of the general government jumped from about 3 percent of GDP in 1970 to 24 percent in 1973. Increasingly, large fiscal deficits were financed with money creation, which translated into galloping inflation despite generalized price controls. By the end of 1973, the country was in a deep economic crisis, with GDP contracting by about 6 percent and end-year inflation exceeding 550 percent.

Fiscal Policy Reforms: Fiscal Consolidation and Comprehensive Public Sector Reform

A. Overview1,2

4. In late 1973, the military took over and adopted an adjustment program aimed at drastically reducing the state’s role in the economy (Reform episode I: 1974–79). The new leadership enacted a radical overhaul of economic institutions and policies, with the objective of shifting from a highly-regulated and closed economy dominated by state property and government control, to a deregulated and open market economy based on private ownership. Cornerstones of the 1974–79 program were a sharp tightening of fiscal policy by cutting primary spending (particularly on wages, transfers, and investment) and adopting radical structural reforms, including the introduction of a value added tax (VAT) and privatization of a large number of state-owned enterprises (SOEs). These measures were part of a package of broader reforms which included most notably the removal of price controls, trade liberalization, and financial markets’ deregulation (Table 2).

Table 2.

Chile: Major Fiscal Reforms

article image
Source: Corbo (1985), IMF, and Mackenzie and others (1997a).

5. This program suffered from some policy inconsistencies, however, that contributed to the build-up of large imbalances and prompted further reforms in 1983 (Reform episode II, 1983–89). In the late 1970s, Chile adopted an exchange rate-based program to reduce stubbornly high inflation, while maintaining full indexation of most contracts (including wages) to past inflation. This generated a sharp real appreciation of the peso and a drastic reduction in the cost of foreign borrowing, which resulted in large capital inflows and unsustainable trade deficits. In 1982, the sudden stop in external financing sparked by the international debt crisis and growing doubts about the fixed exchange rate policy triggered a severe recession and a banking crisis.3 To restore macroeconomic stability, the government adopted a new stabilization program in 1983. This second reform program focused on fiscal retrenchment supported by a wide range of structural measures, such as: further privatizations and a reform of the social security system; labor market flexibility; financial system reform; adoption of a flexible exchange rate regime; and enhanced central bank independence. The reform effort was supported by financial arrangements with the IMF, the World Bank, and the Inter-American Development Bank (IDB).4

6. Both reform programs featured large fiscal adjustments mainly focused on primary spending cuts (Table 2). During the first reform episode (1974–79), the general government’s primary fiscal balance posted an unprecedented correction, reflecting a dramatic compression of primary spending and comprehensive measures to broaden the tax base. Such an unprecedented adjustment was in part motivated by the severity of the economic recession in 1975. During the second reform episode (1983–89), fiscal adjustment was less pronounced as fiscal space created by further primary spending cuts was used to finance a reduction in the tax burden—particularly on business income—and an increase in public investment.

7. Initial spending reforms aimed at achieving immediate and permanent savings, leaving limited room for pro-growth measures. Between 1974 and 1979, capital expenditure was severely reduced—particularly for public housing programs. The wage bill was cut through wage restraint and reductions in employment. Consumer subsidies and public transfers also decreased significantly as thousands of prices were liberalized and more than ⅔ of SOEs were privatized by end-1979. At the same time, health and education expenditures were protected to the extent possible after initial severe cuts (Figure 7), allowing Chile to maintain relatively high health and education standards compared to peers (Mackenzie and others, 1997a; Meller, 1992). During the second reform episode, further reductions in public transfers and subsidies helped create space for a recovery in public investment to alleviate increasing bottlenecks in basic infrastructure, such as roads.

Figure 2.
Figure 2.

Chile: Real GDP Growth, 1970–2014

(Percent, annual)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: WEO.Note: shaded blue area indicates reform period; orange area indicates global financial crisis.
Figure 3.
Figure 3.

Chile versus Emerging Market and Middle-Income Latin America

(Annual GDP growth, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Emerging market and middle-income Latin America (excluding oil producers and Dominican Republic).
Figure 4.
Figure 4.

Synthetic Control Method: Chile

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.Note: Shaded blue areas indicate reform periods.1/ Uruguay, Peru and Colombia.2/ Uruguay, Peru and Colombia.3/ Uruguay and Colombia.4/ Colombia, Uruguay and Peru.
Figure 5.
Figure 5.

Chile: Growth Contributions, 1974–1999

(Percent, annual)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Penn World TableNote: shaded blue areas indicate reform period.
Figure 6.
Figure 6.

Chile: Growth Transmission Channels

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

8. Tax policy reforms led to a radical overhaul of the tax system, including the early introduction of a VAT. In 1975, the cascading sales tax was replaced with a broad-based VAT, featuring a general rate of 20 percent and limited exemptions for specific goods. This helped reduce the distortive effects of sales taxes and excises on relative prices of consumer goods, and yielded substantial revenue thanks to its simple design. After initial increases, income taxes were adjusted to reduce the dispersion of effective tax rates (e.g., the top PIT rate was reduced by two percentage points by 1979); and trade liberalization implied the introduction of a uniform import tariff of 10 percent and the elimination of quantitative restrictions (Mackenzie and others, 1997a). During the second reform period, spending reforms created space to reduce further corporate taxes and lower the standard VAT rate.5

9. Continued improvement in revenue administration and public financial management (PFM) systems helped enact fiscal reforms swiftly. Chile entered the first reform episode with reasonably well-functioning revenue administration and PFM systems, which permitted to enhance tax collection, and to exert greater control on budget preparation and execution (Mackenzie and others, 1997a). During 1974–79, main improvements in tax administration included greater taxpayer education, simplification of taxpayer forms, and streamlined filing and payment procedures. In the PFM area, notable advances comprised tighter control over the budget preparation process; enhanced cash management system; and the adoption of a medium-term macroeconomic and planning framework for investment projects (Mackenzie and others, 1997a; Kalter and others, 2004).6 During the second reform period (1983–89), the government established a copper stabilization fund (CSF) to shield the conduct of fiscal policy from the volatility of copper’s international prices.

B. Notable Design Features

10. Initial reforms were part of a comprehensive, crisis-driven policy package that included some internal inconsistencies. During the first reform episode, the policy package ranged from dramatic fiscal stabilization measures to radical structural reforms—including an overhaul of the tax system, large-scale privatizations, market and price liberalizations, and trade and financial reforms. The breath of these policies was dictated primarily by the gravity of the crisis rather than by considerations on their conduciveness to growth or their impact on the income distribution. As a result, the policy mix included some internal inconsistencies, such as the coexistence of full wage indexation with a preannounced nominal exchange rate (Corbo and Fischer, 1993). During the second reform episode, consistency and synergies among policies were enhanced.

11. Military government and strong institutions facilitated a swift implementation of comprehensive fiscal reforms. While military leadership facilitated the swift identification, adoption and implementation of the far-reaching reforms (Schmidt-Hebbel, 2008), Chile’s relatively good tax and PFM systems were instrumental in carrying out complex reforms in a timely fashion (Mackenzie and others, 1997a). Notable examples are the successful introduction of a VAT in 1975 and the effective reduction of unproductive spending during both reforms episodes, including by rationalizing and better selecting investment programs.

12. Reform packages, while aiming primarily at attaining macroeconomic stabilization, included some measures to mitigate the impact of adjustment on the very poor. For example, at the outset of the first reform period, the government implemented a temporary public works program to alleviate the impact of the steep increase in high unemployment (Edwards and Edwards, 2000); and increased food provisions for participants in the public works programs (Mackenzie and others, 1997a). Also, it introduced other poverty alleviation schemes that targeted mothers, pregnant women, and school children (Meller, 1992).

C. Outcomes

13. While volatile in both reform periods, growth picked up markedly in the late 1980s and remained sustained during the 1990s (Table 4). During 1974–75, GDP fell cumulatively by about 12 percent due to significant terms-of-trade shocks and policy-related factors, such as a sharp fiscal contraction and measures to contain inflation after price liberalization.7 Similarly, economic activity contracted by about 16 percent in 1982–83 as a result of a further worsening in the terms of trade and expenditure-based fiscal retrenchment in the midst of the international debt crisis (Ffrench-Davis, 2002). Growth started to rebound in the mid-1980s peaking at 11 percent in 1989 and hover at around 6 percent in the following decade as improved policy coordination and business-friendly policies supported strong private investment and export performance (Goldsbrough and others, 1996). While such a rebound may partially reflect a cyclical recovery from the 1982–83 crisis, improvements persisted for long enough to suggest that other influences— including fiscal and market-oriented structural reforms—played a significant role (Mackenzie and others, 1997a).

Table 3.

Chile: Economic Growth Predictor Means before Fiscal Reform

article image
Source: IMF Staff calculations.

All variables except GDP growth and GDP growth per capita are averaged for the 1961–1973 period. The average for GDP growth and GDP growth per capita is calculated using the years 1961, 1967 and 1973.

All variables except GDP growth and GDP growth per capita are averaged for the 1961–1982 period. The average for GDP growth and GDP growth per capita is calculated using the years 1961, 1971, 1982.

Table 4.

Chile: Selected Indicators

(Period averages)

article image
Source: Banco Central de Chile, DIPRES, GFS, Haver Analytics, IFS, Penn World Table, SWIID 5.0, WEO, and WDI.

Growth accounting estimates are calculated using Penn World Table 8.0.

From 1960–1990, figures are from Indicadores Económicos y Sociales de Chile. From 1991–2013, figures are from DIPRES. This is true for all revenue, expenditure, and resulting balance figures.

DIPRES data from 1991-present does not break out tax revenue into direct and indirect taxes. For these years, figures are estimated by applying the growth rate in indirect tax revenue (data from Central Bank authorities) to the split between direct and indirect tax revenue in 1990.

Estimates of current and capital expenditure are calculated using shares obtained from GFS.

Gross central government debt.

From 1970–1979, figures are from Edwards and Lustig (1997). From 1980–2012, figures are from CEPAL

Percent of the entire population that is employed.

“Market” Gini measures the income distribution before taxes/transfers; “net” refers to after taxes/transfers.

SWIID reports data in centered five year moving averages. For intervals above that are not five years long, an average of the two middle values is reported.

Percent of the population living on less than $2 (PPP) per day.

14. Sharp expenditure-based fiscal adjustment helped put public finances on a sustainable footing, while making room for some pro-growth fiscal measures down the road. During the first reform episode, the primary fiscal balance of the general government moved from a deficit of about 23 percent of GDP in 1973 to a surplus of about 5 percent in 1979. Two-thirds of the adjustment came from a dramatic reduction in primary spending, with primary current expenditures and public investment falling by 20 and 4 percentage points of GDP, respectively. At the same time, revenues increased by 4 percentage points of GDP following the introduction of a VAT. During the second reform period (1983–89), the improvement in the primary balance was somewhat smaller, about 9 percentage points of GDP, with current primary spending falling by about 8.5 percentage points of GDP. However, capital spending was increased marginally and the tax ratio was lowered by about 3 percentage points—including by reducing CIT and VAT rates.8

15. Fiscal and structural reforms succeeded in paving the way for private-sector-led growth, which in turn helped sustain the fiscal consolidation. Initial efforts to reduce primary public expenditure were further supported by a massive program of privatizations, which brought the number of state-owned enterprises to 38 in 2000, from 595 in 1973 (Serra and others, 2001).9 At the same time, the 1981 pension reform—which included a shift to privatized savings plans and an increase in the retirement age—helped boost private savings and deepen capital markets, which along with significant reductions in CIT rates promoted a sharp increase in private investment throughout the 1990s. These measures ultimately improved allocative efficiency and expanded the economy’s productive capacity, with ensuing medium-term benefits for the budget (in the form of lower subsidies and higher revenue collection). Importantly, the full benefits of such reforms were grasped only after the removal of wage indexation at the end of 1982.

16. Despite some efforts to protect the poor, income inequality increased sharply during the reform periods and remained above the 1973 level throughout the 1990s (Figure 7). Amidst strong growth and the authorities’ efforts to protect the poor, the poverty rate fell to about 12 percent in 1992 from about 21 percent in 1970 (Camhi and Castro, 2003). However, income inequality, as measured by the GINI coefficient, increased from about 0.45 in 1973 to about 0.56 in 1987, mainly reflecting negative labor market developments and depressed real wages (Laban and Larrain, 1995).10 Inequality improved somewhat during the 1990s, but it continued to be on average 20 percent higher than in 1970 (Contreras and Ffrench-Davis, 2012).

Impact on Long-Term Growth

A. Long-Term Growth Dissected

17. Long-term growth picked up significantly in the mid 1980s and was sustained throughout the 1990s (Figure 2). Average annual GDP growth, which hovered around 1.6 percent during the five years preceding the first reform episode, increased to 3 percent during the first reform episode and ultimately to 5 percent during the second reform episode. In the first half of the 1990s, trend growth even accelerated to 7 percent, on average, on the back of buoyant private investment and a rapid expansion of exports, before slowing down in the wake of the Asian crisis (1997–98).

18. Simple comparisons suggest that Chile’s growth performance was stronger than that of its peers following its fiscal reforms (Figure 3). During 1980–89, Chile exceeded the average growth rate of emerging Latin American economies by about 1.7 percentage points. This difference seems to have increased in the decade following the second reform period spanning from 1990 to 1999, with Chile outperforming its peers by about 3 percentage points. More evidence of the 1990s as Chile’s “golden period” in terms of economic growth is presented in several empirical studies, for instance De Gregorio (2004a), and Gallego and Loayza (2002).

19. Similar results are obtained if Chile’s growth experience is compared to a hypothetical growth performance based on the synthetic control method 11(Figure 4 and Table 3). While the results of this methodology need to be interpreted with caution, some general trends seem to emerge. Average GDP growth in the ten years following the first reform episode was about 2.6 percentage points higher than that generated by the synthetic control group based on standard growth predictors.12 Moreover, Chile’s advantage appears to have increased further in the decade following the second reform episode (1990–99). Similar results are reached with the same exercise using GDP per capita growth.13

B. Fiscal Reform: A Game Changer?14

20. Growth accounting suggests that capital and labor were the main drivers of growth after 1974, but there are significant differences across sub-periods (Figure 5). Over the entire sample period, capital and labor recorded an average contribution to growth of 2.3 and 1.8 percent, respectively, while total factor productivity (TFP) only contributed 0.4 percent.15 However, period averages hide significant differences in the relative importance of sources of growth over time. For example, capital accumulation explains half of average growth during the first reform episode, yet TFP was the main contributor to economic development in the years immediately following the 1975 recession. Between 1980 and 1982, capital accumulation was the only source of growth as the contribution of TFP turned negative. In contrast, labor emerged as the main driver of growth during the second reform period, accounting for almost two-thirds of growth.16 Finally, between 1990 and 1999, more than half of average GDP growth was explained by capital accumulation, driven by record high private investment rates.

21. Capital accumulation may have benefited from fiscal measures aimed at restoring macroeconomic stability and promoting private sector development. Fiscal retrenchment and privatizations led to an environment more conducive to private investment by reducing the state’s role in the economy, improving borrowing conditions, and anchoring markets’ expectations on future policies (Corbo, 1985; Corbo and Fischer, 1993; Ffrench-Davis, 2002). In the 1990s, additional factors may have contributed to the significant increase in capital accumulation, such as a drastic reduction in corporate income tax rates (Vergara, 2004) and increased public spending on infrastructure, with potential crowding in effects for private investment (Ffrench-Davis, 2002). In addition, more effective PFM systems may have helped select and implement investment in projects with a higher growth dividend.

22. Improved TFP likely benefited from measures to reduce the state’s role in the economy and to strengthen fiscal institutions. During 1976–79, TFP gains may have reflected the efficiency improvements of economic deregulation, including price liberalization and privatization (Coeymans, 1999). The significant reduction in inflation also seems to have played an important role in enhancing resource allocation (Jadresic and Zahler, 2000). In the late 1980s, aggregate productivity gains may have been related to the adoption of comprehensive labor and pension reforms, which further helped to direct resources to the most productive activities (Edwards and Edwards, 2000). Also, well-functioning tax and PFM systems might have helped to raise the productivity of public investment and to crowd in foreign direct investments, with associated knowledge transfers (Bergoeing and others, 2010).

23. Also, the sharp increase in the contribution of labor in the late 1980s may have been underpinned by macro stability, pro-growth tax and spending policies, and labor reforms. Specifically, fiscal retrenchment promoted private sector activity by reducing uncertainty associated with fiscal imbalances and improving financing conditions (Mackenzie and others, 1997a), which eventually supported job creation. In addition, labor demand and supply benefited from the reduction in income tax rates (particularly on business income) and social security contributions in the context of the 1981 pension reform (Edwards and Edwards, 2000). Additional stimulating factors could have been the substantial restructuring of the civil service and removal of full wage indexation in 1982, which increased real wage flexibility in the face of record high unemployment (Corbo, 1985; Corbo and Fischer, 1993). Also, the allocation of more budgetary funds to the education system could have reduced labor mismatches by increasing workers’ skills. Importantly, the implementation of the above measures appears to have been facilitated by well-functioning PFM systems (Mackenzie and others, 1997a).17

24. In sum, fiscal policy appears to have played an important role in boosting Chile’s growth performance after 1973 by promoting productive factor accumulation and some gains in TFP (Figure 6). Fiscal consolidation contributed to restoring macroeconomic stability and creating the conditions for durable private sector development—including by implementing difficult tax and expenditure reforms with the support of well-functioning fiscal institutions. Also, ambitious labor and pension reforms as well as large-scale privatizations were important in reducing the state’s role in the economy, improving resource allocation, and promoting private investment. However, fiscal policies were not the only source of change, of course, and other external and internal factors also played a key role.

Figure 7.
Figure 7.

Chile: Fiscal Policy and Growth, 1960–2013

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: CEPAL, IMF staff calculations, OECD, Penn World Table, SWIID 5.0, and WEO.Note: Shaded blue areas indicate reform periods.

Lessons

25. Large fiscal retrenchment for stabilization purposes may impose short-term pain, but is not at odds with achieving higher long-term growth. As shown in Chile’s case, comprehensive and growth-friendly tax and spending reforms can ultimately contain the pain associated with fiscal retrenchment by making the tax structure more growth friendly (e.g., by moving from direct to indirect taxation) and improving the quality of spending (e.g., by reducing unproductive expenditures). Chile’s experience also confirms that those reforms that take time to produce net fiscal savings (e.g., tax reform, civil service reform) should start early on in the adjustment process; and that initial spending cuts can be restored at least partially once stabilization takes hold and revenue-raising measures start bearing fruit.

26. However, inconsistencies with other macroeconomic policies, if unaddressed, can jeopardize the potential benefits of growth-oriented fiscal adjustment. Chile’s performance in 1978–82, when exchange rate stabilization programs coexisted with full wage indexation, is emblematic in this regard.

27. Strong and effective fiscal institutions are critical to bring about profound change and maximize the benefits of fiscal reforms. Well-functioning PFM and revenue administration systems permitted the enforcement of substantial changes in the composition of spending and revenues in a timely manner; while, higher revenue collection helped create fiscal space for growth-enhancing policies.

28. Fiscal adjustment strategies mainly aimed at boosting growth can face major challenges with respect to income inequality. Chile’s experience confirms that adjustment strategies that successfully boost growth and reduce poverty do not necessary lead to durable improvements in income inequality—and may even have a detrimental impact on the income distribution. This underscores the importance of specific and well-targeted measures that not only protect the most vulnerable groups, but also ensure that all citizens share equally in the benefits of higher growth.

Appendix. Robustness Tests for Chile

A. Placebo Tests

1. Each country in the group of the input group used was alternatively chosen as the tested country when in fact no fiscal reform took place in that country (Figure A1). If the estimated effect for Chile is significantly larger than the placebo countries, it strengthens the case that the growth gap estimate in the baseline can indeed be attributed to the fiscal reform. Post-reform growth in Chile was higher relative to the growth distribution of non-reformers (countries in the comparator group), in terms of the size of growth gaps. However, the ratios of post-reform and pre-reform RMSPEs are not high, as the synthetic series for Chile were not able reproduce the pre-intervention outcome variable very well. This indicates that the positive growth gap in the post-reform needs to be cautiously interpreted.

A1.
A1.

Placebo Test: Chile

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.Note: Shaded blue areas indicate reform periods.1/ Emerging market and middle-income Latin America (excluding oil producers and Dominican Republic).2/ Emerging market and middle-income Latin America (excluding oil producers and Dominican Republic).

B. Sequential Exclusion of a Comparator Country

2. The sensitivity of the baseline results was tested by sequentially excluding each of the countries in the comparator group that received a positive weight and re-estimating the model (Figures A2 and A3). This implied the sequential exclusion of the following countries for both reform periods: Colombia, Peru and Uruguay. The post-reform growth gap is higher than the pre-reform growth gap in all cases, confirming the baseline.

Figure A2.
Figure A2.

Sequential Exclusion of a Comparator Country: Chile, 1974

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Uruguay, Peru and Brazil.2/ Uruguay and Colombia.3/ Peru, Colombia and Argentina.
Figure A3.
Figure A3.

Sequential Exclusion of a Comparator Country: Chile, 1983

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Uruguay, Peru and Argentina.2/ Uruguay and Colombia.3/ Argentina.

C. Changes in the Reform Timing

Changes in the start year of reform

3. Synthetic series were re-estimated for the years 1969, 1973, 1975 and 1979 (first period) and 1978, 1982, 1984 and 1988 (second period) (Figure A4 and A5). If the growth gap estimate in the baseline changes substantially in response to a slight variation of the starting year, it would cast doubt on the existence and the size of the positive growth gap. If the growth gap does not change substantially in response to a large variation in the starting year, it would undermine our confidence that the positive growth gap is indeed indicative of the effect of the fiscal reform. Results show that the growth effects in the baseline are not sensitive to small changes in starting periods (1973 and 1975 for the first period and 1982 and 1984 for the second period), but setting the start of reform in 1969 results in disappearance of the growth effects. These results are consistent with the conjecture in the baseline that the fiscal reform is the likely reason for the positive growth effect in the baseline. On the other hand, the exercise showed positive growth effect that 1978 and 1979, 1988 also showed a positive growth gap.

Figure A4.
Figure A4.

Change in Reform Period-Start Year: Chile, 1974

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Peru, Colombia and Uruguay.2/ Peru, Uruguay and Colombia.3/ Uruguay, Peru and Colombia.4/ Colombia, Uruguay and Argentina.
Figure A5.
Figure A5.

Change in Reform Period- Start Year: Chile, 1983

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Colombia, Argentina, Peru and Uruguay.2/ Colombia, Uruguay and Peru.3/ Uruguay, Peru and Colombia.4/ Colombia, Uruguay, Peru and Argentina.

Changes in end year of reform

4. The difference in average growth rates was recalculated by setting the end of the fiscal reform period to 1982 and 1984 (first period); and to 1992 and 1994 (second period), respectively (Figure A6 and A7).The purpose of this exercise is to examine whether our conclusion is sensitive to the timing of the end of the reform, given its uncertainty. The difference in average growth rates was recalculated by setting the end of the fiscal reform period to different years. The purpose of this exercise is to examine whether our conclusion in the baseline is sensitive to the timing of the end of the reform, given its uncertainty. Results confirm that the conclusion holds for both 1974 and 1983 reforms when the end of the fiscal reform period changes.

Figure A6.
Figure A6.

Change in Reform Period-End Year: Chile, 1974

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Uruguay, Peru and Colombia.
Figure A7.
Figure A7.

Change in Reform Period – End Year: Chile, 1983

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Uruguay and Colombia.

Germany: Fiscal Policy and Long-Term Growth

A. Background

1. By the late 1990s and the early 2000s, Germany was often referred to as ‘the sick man of Europe’ given its weak growth performance and poor unemployment record relative to its peers. Between 1993 of 2002, Germany’s growth was consistently below the EU average, and in 2002 and 2003, Germany’s growth was even trailing that of almost all other EU and OECD countries (Figure 1). The weak growth performance was mirrored by rising unemployment from 5.5 percent in 1992 to 11.3 percent in 2005.

Figure 1.
Figure 1.

Germany: Pre-Reform Economic Indicators, 1991–2002

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Haver Analytics, Penn World Table, and WDI.

2. By 2003, the fiscal balance had worsened significantly. A fiscal deficit widened from around 1 percent in 2000 to a peak of over 4 percent of GDP in 2003, and by 2005, Germany’s deficit had exceeded the 3 percent threshold laid down in the Stability and Growth Pact several years in a row. While cyclical effects – the German economy stagnated in 2002 and was in recession in 2003 – certainly contributed to undermining revenue performance and to rising social expenditure, there was also long-term deterioration of the fiscal stance. The debt ratio rose from around 41 percent of GDP in 1991 to 68 percent of GDP in 2005, and on average, the fiscal deficit was around 2.5 percent of GDP during the same period. Key factors that contributed to the deterioration of the fiscal position included large spending programs in Eastern Germany following German reunification in 1990, the persistently weak economic performance and the burden associated with liabilities inherited from the former Eastern Germany.

Fiscal Policy Reforms: Expenditure-Based Consolidation and Labor Market Reforms

A. Overview

3. In 2003, Germany embarked on an ambitious and wide-ranging reform program commonly referred to as “Agenda 2010” (Table 1). The reforms were designed to fix several key structural economic problems underlying Germany’s weak long-term economic performance. They included the introduction of more flexible hiring and firing rules, a massive modernization and reorganization of the labor office, which were aimed at facilitating job placements, a reduction of the duration and eligibility of unemployment and welfare benefits, and the slashing of non-wage labor costs through reduced health insurance premiums. One central motivation was to reform and liberalize the labor market to boost both labor supply and labor demand.1

Table 1.

Germany: Major Fiscal Reforms

article image
Source: Cheptea and others (2014), Dullien (2006), IMF, and WEO.

4. The reforms were also motivated by fiscal concerns and had an important fiscal dimension. Against the backdrop of significant fiscal deterioration and the breach of the Stability and Growth Pact, expenditure-based consolidation was another important objective. Following the implementation of the reforms, public spending dropped by almost 5 percent of GDP from its peak at around 47 percent of GDP until 2007 which was especially driven by declines in the wage bill and social spending (amounting to 3.7 percent of GDP). Accordingly, the measures that had important immediate implications for public spending included in particular the reforms affecting unemployment and welfare benefits. Moreover, the potential indirect effects on public spending – through a reduction in unemployment and stronger growth – also helped to contain social spending.

5. The savings on the expenditure side were large enough to create fiscal space that allowed for tax cuts. The top marginal income tax rate was cut from 48 percent to 42 percent in 2005 (which was however partially reversed in 2007), whereas the marginal tax rate for low incomes was cut from approximately 20 percent to 15 percent between 2003 and 2005. In 2008, the federal corporate tax rate was further reduced from 25 percent to 15 percent as part of a larger corporate tax reform that also entailed base broadening as various possibilities for tax deductions were eliminated. While some of the potential demand-side effects were undone by the increase of the standard VAT rate from 16 percent to 19 percent in 2005, the tax system became in any case much less distortionary.

6. After a brief deterioration of the fiscal stance in 2009 and 2010 due to the global financial crisis, these gains in terms of consolidation were ‘locked in’ through a new fiscal rule referred to as ‘debt brake’. The debt brake requires the structural deficit of the federal government not to exceed 0.35 percent of GDP, and that the structural deficit of the German states remains at zero. It will be phased in over the 2011–2020 period and becomes fully binding for the federal government from 2016 onwards. While Germany had already been subject to the Stability and Growth Pact, the debt brake is much more binding as it is included in the German constitution. The debt brake is also much more effective in limiting debt compared to the golden rule that had already been part of the constitution. The introduction of the debt brake may have therefore been instrumental in that the progress towards lowering public deficits has not been reversed in the years after the global financial crisis.

B. Notable Design Features

7. At the onset of the reform episode, the government targeted key rigidities of the labor market and non-wage labor costs, although the immediate fiscal benefits of these reforms in terms of expenditure savings were likely another important motivation. Economically, this type of prioritization was sensible in this situation as the potential growth dividend of structural reforms will only slowly unfold over time and may take many years to materialize. In addition, boosting medium- and long-term growth is an essential precondition for fiscal consolidation to be ultimately successful. As the examples of several European countries are showing, fiscal consolidation targets are very difficult to achieve when potential growth is low, in particular, when the economy is shrinking (Abbas and others, 2012).

8. While the reforms were controversial given their far-reaching changes in social policy, they gained broad support in Parliament. The reforms were initiated by a coalition government of the social democrats, and the green party, implying that the opposition at that time consisted virtually only of the conservative party and the liberal party, which were both sympathetic with the general impetus of the reforms.

9. The sequencing and prioritization of reform efforts helped to safeguard political support. Over time, political capital for reform is likely to wane as voters become ‘tired’ of relentless reform efforts. Therefore, it is important to target the most important impediments and implement the most difficult reforms in political terms first. Indeed, cuts in long- established health benefits and changes in job security, which were not only sensitive issues for the majority of voters, but also critical from an economic perspective, were implemented in the beginning of the reform episode.

10. In addition, political support was not further undermined by additional fiscal consolidation measures. Quite to the contrary, cuts in personal income taxation in 2005 that roughly coincided with the reforms but that were agreed upon in 2000 already may have been important to safeguard voters’ support. In addition, the government concentrated on economic reforms with the biggest ‘bang for the buck’ rather than implementing less important structural reforms that still induce economic pain. However, favorable economic conditions in the mid 2000s that were in part driven due to factors outside of government influence may also have helped to safeguard political support.

11. While from an economic perspective, it would have been desirable to implement the reforms earlier, from a political economy perspective, the timing of the reforms was probably not a coincidence. One factor that triggered the reforms may have been the persisting economic stagnation despite the global economic recovery in the early 2000s which in turn resulted in widespread pessimism about Germany’s growth prospects (Bornhorst and Mody, 2012). At the same time, it may have not been feasible to implement the structural reforms earlier because scarce political capital required to implement them was tied up in issues relating to reunification throughout the 1990s (Kastrop, 2013).2

C. Economic Performance after the Reform

12. Germany’s economic performance since the implementation of the reforms has certainly been remarkable, apart from a relatively deep recession in 2009. Germany recorded strong growth of significantly above 3 percent in 2006 and 2007. Following the global economic crisis, Germany showed a relatively strong growth performance with average growth rates of 2.2 percent between 2010 and 2013 whereas its potential growth is estimated to be 1–1.5 percent. While such a growth record is usually considered as mediocre, by European standards during that period, it was relatively strong. However, Germany is especially praised for its progress towards reducing unemployment: the unemployment rate was almost halved between 2004 and 2013 and reached historically low levels. In the light of the fiscal challenges in other countries in Europe, the improvement in the fiscal consolidation was also impressive, despite a temporary deterioration of the fiscal position during the global financial crisis. An overall deficit of 4.2 percent in 2003 had turned into a fiscal surplus of almost 0.2 percent by 2013, which is projected to persist in 2014 and 2015.

13. After its 10th anniversary, the reform program is therefore widely believed to have been largely successful. Against the background of significant decreases in unemployment, this seems plausible, given that the wide-ranging reforms in the mid 2000s of the ‘Agenda 2010’ targeted the labor market. Measures of the reform packages that are routinely believed to have been especially important include the introduction of more flexible labor contracts and stronger incentives for the unemployed to speed up their job search.

14. However, the contribution of the reforms to the economic resurgence of Germany remains nevertheless controversial. First, wage restraints played an important role in increasing the competitiveness of the German economy. In turn, wage restraints were facilitated by Germany’s highly decentralized system of wage bargaining at the regional, industry or even firm level, and, related to this, by the fact that negotiations are far more often consensus-based compared to other countries (Dustmann and others, 2014). Second, another, important explanation for the muted effects of the global financial crisis on employment may be related to subsidized short-time work arrangements and firm-level working time accounts which both allowed firms to cut working hours per worker rather than the number of workers. Having retained the workers then enabled firms to quickly respond to rising labor demand from 2010 onwards (Burda and Hunt, 2011).

Impact on Long-Term Growth

A. Long-Term Growth Dissected

15. Evaluation of Germany’s growth performance needs to be placed in a proper context, especially given the GFC that took place post-reform (Figure 2). Late 1990s to early 2000s saw a sharp slowdown in growth. The pickup in growth in 2005–07 was followed by a negative growth in 2008–09 and subsequent recovery. Output growth is qualitatively similar in many industrial economies, however, and a more formal approach is needed. The GFC led to abnormally-large output fluctuations, which adds to a difficulty in discerning the long-term trend post-reform.

Figure 2.
Figure 2.

Germany: Real GDP Growth, 1991–2013

(Percent, annual)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: WDI, WEO.Note: shaded blue area indicates reform period; orange area indicates global financial crisis.

16. Simple comparisons of post-reform growth in Germany with its peers indicate that Germany’s growth is relatively favorable post-reform (Figure 3). While Germany’s growth lagged its peers by about 2 percentage points in the 10 years before the reform, post-reform it has been higher by about 0.5 percentage points.

Figure 3.
Figure 3.

Germany versus European Union-15

(Annual GDP growth, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Sources: IMF staff calculations.1/ European Union-15.

17. Counterfactual simulations using the SCM confirm that the fiscal reform has resulted in the growth acceleration.3 Depending on whether we use GDP growth or per capita GDP growth, fiscal reform in Germany has resulted in growth acceleration of about 0.7–1.5 percent (Figure 4 and Table 2).

Figure 4.
Figure 4.

Synthetic Control Method: Germany

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.Note: Shaded blue areas indicate reform periods.1/ Belgium, Austria, United Kingdom, Netherlands and France.2/ Belgium, France, Austria, Greece and Luxembourg.
Table 2.

Germany: Economic Growth Predictor Means before Fiscal Reform

article image
Source: IMF staff calculations.

All variables except GDP growth and GDP growth per capita are averaged for the 1990–2002 period. The average for GDP growth and GDP growth per capita is calculated using the years 1990, 1996 and 2002.

B. Fiscal Reform: A Game Changer?

18. Examining post-reform developments of the factors of production, it appears that labor was the main channel that the fiscal reform affected growth outcome. The growth accounting shows that labor and TFP were the main drivers of the post-reform growth performance (Figure 6, Table 3). While developments in labor can be traced to fiscal and labor market reform, the TFP might have been mainly affected by other factors.

Figure 5.
Figure 5.

Germ any: Grow th Transmission Channels

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Figure 6.
Figure 6.

Germany: Fiscal Policy and Growth, 1991–2013

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Bundesagenturfeur Arbeit, Haver Analytics, IMF staff calculations, OECD, Penn World Table, SWIID 5.0, and WEO.Note: Shaded blue areas indicate reform period; orange areas indicate global financial crisis.
Table 3.

Germany: Selected Indicators

(Period Averages)

article image
Source: Deutsche Bundesbank, Havery Analytics, OECD, Penn World Table, SWIID 5.0, WEO, and WDI.

The figures listed under “2011–13” for real GDP, growth contributions, and poverty reflect 2011 data only.

Growth accounting estimates are calculated using Penn World Table 8.0.

ESA 95 accounting.

Percent of the working age population (15–64) that is employed.

“Market” Gini measures the income distribution before taxes/transfers; “net” refers to after taxes/transfers.

Proportion of the population falling below the poverty line, which is defined as half the median household income after taxes/transfers.

19. The above chart identifies potential transmission channels from fiscal policy to growth (Figure 5). Spending reform, such as streamlining unemployment benefits and pension reform, increased incentives for job search and contributed to an increase in labor supply. The spending reform also created fiscal space for a reduction of the tax burden, which contributed to an increase in labor supply as well. Finally, labor market reform made job matching more efficient, and thus increased the level of job creation for given levels of labor supply and demand.

20. Superior employment performance to its peers reflects the effect of comprehensive package of fiscal and labor market reforms.4 The continuous contribution of labor reflects structural changes that moderated wages, increased steady state employment, and improved matching in the labor market. In addition, enhanced flexibility of the labor market enabled cyclical responses to the severe negative shock without massive layoffs.5

21. Developments of the TFP may be explained by factors other than the fiscal reform: the GFC and product market reform. After a negative growth in the first half of the 2000s, there was an uptick in the TFP growth from 2006–07, with notable contributions by the services/financial and business services sector. (Poirson, 2013). The TFP sharply decreased in 2008–09 in the face of the GFC, from which it recovered sharply and returned to a positive growth path. Product market reform may have been a contributing factor (OECD, 2014).

22. Capital accumulation was subdued. Capital contribution to growth has seen a gradually decreasing trend. Both public and private investment to GDP ratio continues to be lower than euro area average. (IMF, 2014d) There is no clear evidence that fiscal and labor market reform affected investment.

23. Income inequality saw a slight increase until 2007, both before and after redistribution, and has largely stabilized afterwards. The market Gini coefficient may have been affected by changes in capital income and greater labor income dispersion. Additionally, the net Gini coefficient was also affected by the weakening of the tax and public transfer system, due to the decrease in top income tax rates, and replacement of public transfer by labor income that accompanied a trend increase in employment (Schmid and Stein, 2013).

24. Summing up, there is evidence that fiscal and labor market reforms may have contributed to an increase in long-term growth by facilitating strong employment growth. However, it should be noted that potential growth dividend of structural reforms only slowly unfolds over time and may take many years to materialize. Given the interruption from the global financial crisis, the full effects of structural reforms are yet to be seen, including indirect effects through the increase in inequality.

Lessons

25. Reforms that aimed at promotion of long-run growth were helpful in sustained growth and fiscal consolidation. Despite a temporary deterioration of the fiscal position during the global financial crisis, fiscal consolidation over the last 10 years has been very successful. In this context, the central lesson from Germany is that consolidation measures should ideally result in spending cuts and in the promotion of long-run growth. The reforms in Germany targeted key rigidities of the labor market and entailed cuts in unemployment benefits. The rationale for this strategy is that in case of a prolonged recession, improving the fiscal position is very difficult: debt as a share of GDP increases if the economy shrinks even in the absence of deficits, and lowering deficits is considerably harder with automatic increases in social spending and automatic decreases in tax revenue in recessions.

26. However, there are signs that the recent strong economic performance and buoyant public revenue made it harder for politicians to safeguards the economic achievements of the reforms. In the area of pensions, recent measures facilitated early retirement in some cases and resulted in higher pensions for specific population groups. Other controversial measures include recently introduced child care-benefits, which have been shown to be a powerful incentive for labor participation by the parents, however can be economically costly. Finally, a uniform minimum is set to become effective which will result in significant wage increases in some regions.

27. The reforms may have contributed to the slight increase in income inequality. Dustmann and others (2014) report wage growth of different percentiles of the wage distribution in western Germany (they argue that similar wage growth decomposition for eastern Germany is not insightful due to the transition after German reunification). They show that real wages at the 15th percentile fell significantly, in particular between 2003 and 2008, whereas wages at the 85th percentile still increased over the same time horizon.

Appendix. Robustness Tests For Germany

A. Placebo Tests

1. Every country in the comparator group1 was alternatively chosen as the tested country when in fact no fiscal reform took place in that country2 (Figure A1). If the estimated effect for Germany is significantly larger than the placebo countries, it strengthens the case that the growth gap estimate in the baseline can indeed be attributed to the fiscal reform.

Figure A1.
Figure A1.

Placebo Test: Germany

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.Note: Shaded blue area indicates reform period.1/ European Union-15.

2. Post-reform growth in Germany was not demonstratively higher relative to the growth gap distribution of non-reformers (countries in the comparator group), in terms of either the size of growth gaps or ratios of post-reform and pre-reform RMSPEs. Thus the tests cannot rule out the possibility that the growth estimate in the baseline could be due to factors other than the fiscal reform. Results of this placebo test should be interpreted with caution, however, because pre-reform match was poor in application to some comparator countries (Figure A1).

B. Sequential Exclusion of a Comparator Country

3. The sensitivity of the baseline results was tested by sequentially excluding each of the countries in the comparator group that received a positive weight and re-estimating the model (Figure A2). This implied the sequential exclusion of the following countries: Austria, Belgium, France, Netherlands and United Kingdom. The post-reform growth gap is higher than the pre-reform growth gap in all cases, confirming the baseline.

Figure A2.
Figure A2.

Sequential Exclusion of a Comparator Country: Germany, 2003

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ France, Netherlands, Ireland and Luxembourg.2/ France and Austria.3/ Austria, Belgium, Netherlands and United Kingdom.4/ France and Austria.5/ Belgium, Austria, France, Netherlands and Greece.

C. Changes in the Reform Periods

4. The synthetic series was re-estimated for the years 1998, 2002, 2004 and 2008 (Figure A3). If the growth gap estimate in the baseline changes substantially in response to a slight variation of the starting year, it would cast doubt on the existence and the size of the positive growth gap. If the growth gap does not change substantially in response to a large variation in the starting year, it would undermine our confidence that the positive growth gap is indeed indicative of the effect of the fiscal reform. Results show that the growth effects in the baseline are not sensitive to small changes in starting periods (2002 and 2004), but setting the start of reform in 1998 results in disappearance of the growth effects. These results confirm that the fiscal reform is the likely reason for the positive growth effect in the baseline.

Figure A3.
Figure A3.

Change in Reform Period-Start Year: Germany, 2003

(Annual GDP growth real versus annual GDP growth synthetic, 10-year average)

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: IMF staff calculations.1/ Austria, France and Luxembourg.2/ Belgium, Austria, France, Greece and Luxembourg.3/ Belgium, Austria, Netherlands, France and United Kingdom.4/ France, Netherlands, Belgium and Sweden.

D. Changes in the Reform Periods

5. Omitted as there are not enough post-reform observations.

Ireland: Fiscal Policy and Long-Term Growth

A. Background

1. Fiscal policy has played both positive and negative roles in shaping the trends in Irish growth. During the 1970s and 1980s the poor design and execution of fiscal policy likely acted as a drag on growth. Likewise, pro-cyclical policies and corresponding structural deficits in the mid-2000s left Ireland ill-equipped to deal with a housing and commercial property crash and global financial crisis. On the other hand, as this case study seeks to highlight, fiscal policy played a contributing role in creating the conditions under which economic growth could flourish during the ‘Celtic Tiger’ period of the 1990s.

2. The 1967 to 1986 period in Ireland was characterized by slow growth, stagnant living standards and a deteriorating fiscal position (Figure 1). Real GDP growth1 averaged 3.8 percent over this period and GDP per capita averaged 62 percent of the level in the EU and improved only marginally. Due to a rapid fiscal expansion and an unfavorable external environment, the fiscal position deteriorated significantly with budgetary deficits averaging 7.6 percent of GDP and the debt-to-GDP ratio doubling to 123 percent of GDP by 1987 from 53.6 percent in 1970.

Figure 1.
Figure 1.

Ireland: Selected Indicators, 1967–2007

Citation: Policy Papers 2015, 023; 10.5089/9781498344654.007.A002

Source: Eurostat, WEO.

3. Building on education and investment measures implemented earlier, the reforms of the period 1987–89, marked the beginning of a remarkable economic expansion. Over the near 18-year period post 1989, real GDP growth averaged 6 percent and living standards, measured using GDP per capita, caught up to and eclipsed the EU average. Primary deficits turned to surplus in 1987, balanced budgets were achieved in 1997 and the debt to GDP ratio declined dramatically to 24 percent by 2007. Various factors contributed to what is often referred to as the “Celtic Tiger.” Fiscal consolidation helped reduce economic uncertainty and because it had the support of social partners, a virtuous cycle of low taxes and increased labor competiveness was created. Low labor costs, a high degree of trade openness, an underutilized labor force that was better educated than the generation before it, and favorable tax rates allowed FDI to flourish.

4. Ireland’s growth performance took place against a backdrop of an improving external environment. The exchange rate environment stabilized when Ireland adopted wider exchange rate bands within the European Exchange Rate Mechanism in the second half of 1993. Ireland was a recipient of EU structural funds and the signing of the Maastricht Treaty in 1992 solidified Ireland’s commitment to the European Monetary Union which sent a positive signal to foreign firms looking for investment opportunities within the EU. Finally, the external environment improved considerably as the United States, one of Ireland’s largest trading partners and a source of significant FDI, emerged from recession in 1991.

5. The”Celtic tiger” period of growth ended abruptly with the bursting of the property bubble in 2007 (Figure 2).2 Property-related tax revenues became and ever increasing share of total revenues throughout the 1990s. This trend accelerated in the mid-2000s as export and FDI driven growth gave way to one driven by credit, real estate (see first chart be