Back Matter

References

  • Carrière-Swallow, Yan, David, Antonio C. and Leigh, Daniel, 2018. “The Macroeconomic Effects of Fiscal Consolidations in Developing Economies: Evidence from Latin America and the Caribbean”, mimeo, International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • David, Antonio C. 2017. “Fiscal Policy Effectiveness in a Small Open Economy: Estimates of Tax and Spending Multipliers in Paraguay”, IMF Working Paper 17/63 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Devries, Pete, Guajardo, Jaime, Leigh, Daniel and Pescatori, Andrea 2011. “A New Action-Based Dataset of Fiscal ConsolidationIMF Working Paper 11/128 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Escolano, Julio, Jaramillo, Laura, Mulas-Granados, Carlos and Terrier, Gilbert, 2014. “How Much is A Lot? Historical Evidence on the Size of Fiscal AdjustmentsIMF Working Paper 14/179 (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Guajardo, Jaime, Leigh, Daniel and Pescatori, Andrea, 2014. “Expansionary Austerity? International EvidenceJournal of the European Economic Association 12: 949968.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gunter, Samara, Riera-Crichton, Daniel, Vegh, Carlos A., and Vuletin, Guillermo, 2017. “Non-linear effects of tax changes on output: A worldwide narrative approach”, IDB Discussion Paper IDP-DP-540 (Washington: Inter-American Development Bank).

    • Search Google Scholar
    • Export Citation
  • Jordà, Òscar, and Taylor, Alan, 2015. “The Time for Austerity: Estimating the Average Treatment Effect of Fiscal Policy,” Economic Journal 126: 21955.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, Estelle X., Mattina, Todd, and Poghosyan, Tigran, 2015. “Correcting “Beyond the Cycle: Accounting for Asset Prices in Structural Fiscal Balances”, IMF Working Paper 15/109. (Washington: International Monetary Fund).

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ramey, Valerie, 2016. “Macroeconomic Shocks and Their Propagation,” Working paper 21978 (Cambridge: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Romer, Christina and Romer, David, 2010. “The Macroeconomic Effects of Tax Changes: Estimates Based on a New Measure of Fiscal Shocks.” The American Economic Review 100: 763801.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, Weonho, Fidrmuc, Jan, and Ghosh, Sugata, 2015. “Macroeconomic Effects of Fiscal Adjustment: A Tale of Two Approaches”, Journal of International Money and Finance 57: 3160.

    • Crossref
    • Search Google Scholar
    • Export Citation
1

We are grateful to Sebastian Acevedo, Roberto Cardarelli, Jaime Carrera, Aliona Cebotari, Valentina Flamini, Frank Fuentes, Roberto Garcia-Saltos, Dmitry Gershenson, Jaime Guajardo, Izabela Karpowicz, Nicole Laframboise, Frederic Lambert, Carola Moreno, Abelardo Pachano, Magali Pinat, Daniel Rodriguez, Pedro Rodriguez, Rene Tapsoba, Iulia Teodoru, Saji Thomas, Joyce Wong, Irina Yakadina, as well as the members of the different IMF country teams for helpful comments; and to the staff of the Joint Bank-Fund Library for their support in locating the historical records cited in the paper.

2

A number of recent papers (Liu et al., 2015; and Yang et al., 2015) have tried to mitigate this problem by building CAPB indicators that attempt to address the influence of asset price fluctuations.

3

For a discussion of the potential pitfalls of the narrative approach see Ramey (2016), Jordà and Taylor (2015), and Escolano et al. (2014).

4

As previously discussed, long-run considerations could include for example, measures aimed at reducing inequality, improving incentives, increasing efficiency or based on a philosophical belief in the benefits of small government (Romer and Romer, 2010).

5

An increase in the VAT rate by 3 percentage points with an estimated revenue yield of 2 percent of GDP was implemented in 1995 with the objective of reducing the fiscal deficit in the context of a loss of confidence in debt markets. In line with Gunter et al. (2017), we consider that this episode was primarily motivated by responding to a fall in confidence and a large capital outflows. Therefore, we do not record it as fiscal consolidation motivated primarily by deficit-reduction and medium-term fiscal sustainability considerations. The 1995 Article IV report (EBS 95/150) states on page 2 that: “Facing a loss of confidence and difficulties in rolling over public sector maturities, in March 1995 the authorities took strong measures to bolster the public finances and provide support to the banking system. … The measures, which were expected to yield over 2 percentage points of GDP in 1995, included a temporary 3 percentage point surcharge on the value-added tax (VAT) rate, which was exempted from revenue-sharing with the provinces, and a wage cut for higher-paid public employees. The aim was to shift the overall balance in the public finances (excluding privatization receipts) from a deficit of 0.5 percent of GDP in 1994 (and a projected deficit of 1.5 percent of GDP in 1995) to a surplus of 0.7 percent of GDP in 1995, with privatization expected to generate an additional 0.8 percent of GDP”.

In 2016, the government announced tax hikes amounting to 2 percent of GDP and expenditure cuts of 0.3 percent (see page 11 of IMF country report 16/346 for a detailed description of the measures). But, these policy actions were offset by tax cuts (export taxes) and increases in pension spending. As a result, the federal primary fiscal deficit remained largely unchanged as a share of GDP in 2016 (IMF country report 17/409 Figure 3 and Table 5).

6

While fiscal policy tightening did occur during 1989–94, it was primarily motivated by restraining inflation and responding to balance of payments crises. We therefore do not record it as fiscal consolidation motivated primarily by medium-term fiscal sustainability considerations. For example, the 1989 Article IV Staff Report (EBS/89/209) explains (p. 54) that “Since August 6, 1989 the new government has been implementing policies designed to return the economy to the path outlined in Bolivia’s medium-term program described in the memoranda attached to the Letters dated June 30, 1988 and March 1, 1989. These policies aim at economic growth of 2.8 percent for 1989 while seeking to limit inflation to 12.5 percent at end year.” An increase in the VAT rate from 10 to 13 percent in March 1992 as well as an increase in the prices of petroleum products were expected to generate revenues in the order of 1.25 percent of GDP (see the 1992 Article IV consultation staff report EBS/92/137, Page 4percent of GDP). But there were expenditure overruns in 1992 relative to program targets, reflecting higher investment spending and an increase in the public sector wage bill (see 1993 Article IV consultation staff report EBS/93/210). Moreover, fiscal measures amounting to 2.5 percent of GDP in 1998 were more than offset by increased expenditures linked to structural reforms. The 1998 Article IV staff report (EBS/98/153) discusses on page 7 the motivations for the authorities’ program for 1998: “Against this background, the authorities’ program for 1998 … aimed to support an increase in economic growth to 4½ to 5 percent with a further reduction in poverty. The program also aimed to limit the inflationary effects of an increase in indirect taxes and reduce inflation to 6% percent in 1998”. The report discusses the fiscal measures, which included an increase in taxes on domestic petroleum products in December 1997. But these measures could not fully offset the fiscal costs of structural reforms (especially the pension reform), which were expected to reach 5 percent of GDP in 1998.

A fiscal consolidation of 1.7 percent of GDP occurred in 2003 with tax hikes amounting to 0.7 percent of GDP and expenditure cuts amounting to 1 percent of GDP. But it was motivated by the need to stabilize the economy after civil disturbances. The 2003 Request for Stand-by arrangement report (EBS/03/37) discusses the motivations for the policy actions: “The economic program for 2003, in support of which the authorities are requesting an SBA, focuses on stabilizing the economy after the recent civil disturbances and resulting financial instability and laying the basis for a return to growth”. In addition, page 9 of the same report lists the key measures adopted in August 2003, which include: i) a new tax code that significantly strengthens the enforcement capabilities of the tax and custom agencies; ii) a tax regularization scheme; iii) a set of modifications to the tax bill (Law 843) that eliminated loopholes, expands tax bases, and limits deductions. Box 2 on page 12 of the 3rd SBA review report (EBS/04/73) presents estimates of the revenue yield of these measures in 2003, which are in the order of 0.7 percent of GDP. The authorities also announced steps to cut expenditures by 1 percent of GDP. As shown in Table 2 on page 25 of the 2006 Article IV consultation staff report (IMF Country Report No. 06/270), these plans were implemented.

7

An adjustment program aimed at reducing very high inflation with 1.5 percent GDP in tax measures and 2.5 percent of GDP in expenditure cuts was implemented in 1990. This episode was determined to be related to cyclical conditions and was not included in the database. According to the 1991 Recent Economic Developments report (SM 91/201) page 26: “In 1990 tax revenue of the Central Administration (excluding earmarked social taxes) increased by 1.5 percentage points of GDP as a result of measures implemented in March”. Same report page 30 states that: “Expenditure (excluding outlays shifted to the expanded Social Security Budget) declined by 10.5 percentage points of GDP in 1990 (see Table 1.5); most of the decline (8 percentage points) reflected the effect of negative real interest payments on domestic debt and the remainder resulted from cutbacks in current and capital transfers to public enterprises and a number of measures related to the administrative reform”. The 1993 Recent Economic Developments report (SM 93/125) sheds light on the motivation of the adjustment program on page 4: “Upon assuming office in March 1990 the new Administration introduced an economic program (the Collor I Plan) that aimed at bringing about a sharp drop in inflation, which had reached 72 percent per month in February”.

A fiscal consolidation consisting of tax increases amounting to 1.3 percent of GDP and expenditure cuts of 0.9 percent of GDP (over three years) was implemented in late 1998/early 1999 in the context of a balance of payments crisis. The 2001 OECD Economic Survey report for Brazil has a description of fiscal measures adopted in 1998–1999 on page 58. These measures were deemed to be motivated by cyclical considerations and were not included in the database.

Moreover, important changes to the tax system were approved and implemented in late 2003 and in May 2004 motivated by need to reduce the deficit, but also to increase efficiency. The motivation for the adjustment is discussed by Corbacho in the 2005 IMF selected issues paper (SM/05/83) and also in the 2004 Central Bank of Brazil (BACEN) Bulletin (pages 78–79). The reforms comprised changes in the base (moving them towards a value-added rather than turnover and also incorporating imports) and in the rates for the COFINS and PIS/PASEP taxes. The reform of the COFINS in particular was significantly revenue enhancing, with collections rising by 0.7 percent of GDP in 2004. But these tax increases were also accompanied by exemptions and reductions in other taxes with an estimated annualized cost of 0.5 percent of GDP (see footnote 5 in Corbacho, 2005). In addition, the 2004 BACEN bulletin shows that public expenditures increased by 0.5 percent of GDP in 2004 (mainly public investment and social benefits), which more than offset the net effect of the tax reforms. Biljanovska (2017) in IMF Country Report 17/216 does not identify any episode of fiscal consolidation in Brazil over the period 2001–2015 using the cyclically-adjusted primary balance approach.

8

In March 2011, cuts to current expenditures amounting to about 0.4 percent of GDP relative to the budget were announced as “… a contribution to moderate inflation in a time when expectations had significantly increased” (see page 16 of the 2012 Informe de Finanzas Publicas). This measure was not included in the database given that it was clearly motivated by economic prospects. In addition, the corporate tax rate was increased from 17 percent in 2010 to 20 percent in 2011, and 18.5 percent in 2012, to help finance reconstruction linked to the February 2010 earthquake. This measure was expected to yield 0.3 percent of GDP in 2011 (see page 6 of IMF Country Report No. 12/267 and page 15 of the 2011 Informe de Finanzas Publicas). Given the nature of the motivation of the tax increase, we also chose to exclude it from the database.

10

A fiscal adjustment consisting of cuts in investment spending and increases in public tariffs was implemented by the new administration that took office in 1990, but was primarily motivated by the need to reduce inflation, as indicated on pages 2 and 4 of the 1991 Article IV staff report (SM/91/204): “The program of the authorities for 1990 envisaged a reduction in the rate of inflation to 24 percent and a moderate overall balance of payments surplus. A rate of growth of real GDP of 3 percent was expected to be consistent with these objectives”. In 1991, revenue measures, including an increase in the VAT rate from 10 to 12 percent in January were also motivated by the need to reduce inflation. Page 5 of the 1991 Article IV staff report (SM/91/204) states that: “At the beginning of 1991 the authorities framed a program that aimed at reducing the 12-month rate of inflation to 22 percent by the end of the year”. See the 1991 Article IV staff report (SM/91/204) page 7 and the 1992 Article IV report (SM/92/93) page 2 for a description of these measures and their impact. A tax reform package taking effect in January 1993, including an increase in the VAT rate from 12 to 14 percent, was more than offset by expenditure increases. See the 1993 Article IV staff report (SM/93/160) pages 4 and 5. A tax reform package approved in December 1995 was once again offset by expenditure increases. See the 1995 Article IV report (SM/95/319) page 5.

11

A tax package with an estimated revenue yield of 0.3 percent of GDP was adopted in 2001 (2001 Article IV consultation report SM/01/190 pages 9 and 10), but was offset by increases in wage expenditures and therefore was excluded from the database. In addition, consolidation measures were adopted in late 2002/early 2003, but seem to have been mostly motivated by economic activity and therefore were also excluded from the database. The 2002 Article IV consultation staff report (SM/03/55) states on pages 7 and 8 that: “Anticipating further economic deterioration in 2003, the authorities began to take corrective measures at the end of 2002”. These measures included the approval of a temporary revenue emergency package in December 2002. All tax increases would be in effect only for 2003 and were expected to yield about 1 percent of GDP in revenues. Moreover, a tax reform with the objective of reducing the fiscal deficit with an expected yield 2½ percent of GDP (see IMF country report 11/161 page 4 and Box 1 on page 9) was approved by congress in early 2012, but was voided by the Supreme Court in the same year (IMF Country Report 13/79 page 4). This reform was also excluded.

12

A new tax code was approved in September 1995 (IMF Country Report 98/45, page 19).

13

Starting in August 1990, the government embarked on an economic reform program (entitled “New Economic Program”) and adopted fiscal consolidation measures to increase tax revenues and cut capital expenditures that led to a reduction in the public sector deficit of more than 1 percentage point of GDP in 1990. In addition, an extension of the base for the 15 percent special surcharge on imports announced in July 1991 was expected to provide revenues equivalent to 1½ percent of GDP and an increase in administered fuel prices would generate revenues in the order of ½ percent of GDP during that year (page 9 of the 1991 Article IV staff report-EBS/91/127). On the expenditure side, Box 1)”. The 2002 Article IV consultation staff report (SM/02/148) presents an estimate of net effect of measures on page 6. We consider that this episode was motivated by prospective economic conditions and exclude it from the database. The large improvement in the fiscal balance observed in 2015 seems to be mostly related to a capital grant from the restructuring of the Petrocaribe debt (see page 7 of the 2015 Article IV consultation staff report IMF Country Report No. 16/342. The 2017 Article IV consultation staff report (IMF Country Report No. 17/254) explains further in footnote 3 on page 6 that: “The gains from the 2015 restructuring of the country’s debt under the PetroCaribe arrangement with Venezuela, bought back at a discount of over 50 percent, are recorded as an above-the-line capital grant in 2015 (3 percent of GDP), boosting the headline fiscal balance for that year.” See also page 16 of the 2016 Informe Explicativo y Politica Presupostaria report.

14

Note that the Ministry of Finance has revised the methodology to calculate public sector debt. Under the new methodology, the debt to GDP ratio in 2004 would amount to 46.9 percent of GDP.

15

In 1989 the government adopted several revenue measures in the context of an SBA-supported program (see 1989 Article IV consultation staff report EBS 89/164, page 11). We did not include this episode in the database because the measures seem to have been motivated by a desire to stabilize the economy. The letter of intent for the 1989 SBA request (EBS/89/161) states on page 1 that: “Twenty days after having assumed office, on August 30, 1988, the Ecuadoran Government, in use of its sovereign faculties, adopted an emergency economic program based on a new economic approach consistent with its political philosophy, which was aimed at stabilizing the economy as a first step toward its subsequent recovery …”. See also Page 44 of the 1989 Article IV consultation staff report (EBS 89/164).

Several fiscal actions were undertaken in 1994 as a response to a decline in international oil prices (especially the tax reform passed in December 1993 and expenditure cuts), as discussed in the 1994 Article IV consultation staff report (EBS/94/88) page 4. This episode was also excluded from the database. Moreover, a number of policy actions were also implemented in 1997. Nevertheless, we decided not to include them in the database as they seem to have been motivated by a desire by authorities to stabilize the economy, as indicated in the 1997 Article IV consultation staff report (SM/97/212) page 14. A revenue package consisting of an increase in an import surcharge and a financial transactions tax was implemented starting in February 1999 with an estimated impact of 1.3 percent GDP. But the measures were motivated by a fall in oil prices in 1998 and lack of external financing, hence we chose to exclude this episode from the database (see section IV for a more detailed description of this episode). A tax reform was adopted in 2001, but the measures were subsequently reversed and therefore not included in the database. Gunter et al. (2017) flag the increase in the VAT rate by 2 percentage points associated with the tax reform as exogenous with a debt-driven motivation. Nevertheless, the VAT rate increase was reversed subsequently. According to the 2001 2nd SBA review staff report (EBS/01/72), the expected net increase in fiscal revenues deriving from policy measures in 2001 would be around ½ percent of GDP (Box 2 on page 15). Nevertheless, the 3rd and 4th SBA review staff report (EBS/01/200) states on page 12 that: “In particular, preliminary data on VAT payments through September indicate that improved collection efficiency will almost fully offset the estimated revenue loss (about US$70 million) following the August 2001 reversal by the constitutional tribunal of the increase in the VAT rate (bringing it back to 12 percent from 14 percent) that had been part of the 2001 fiscal program.”

Moreover, a tax reform with the objective of improving equity and raising the tax to GDP ratio was approved by the constituent assembly in December 2007 with an expected yield of 0.6 percent of GDP (2007 Article IV consultation staff report SM/08/11, page 15), but was its impact was more than offset by increases in expenditure.

In addition, a large fiscal adjustment took place in 2015 with tax measures expected to yield revenue increases in the order of 2.5 percent of GDP, whereas expenditure cuts would amount to close to 3 percent of GDP. But the 2015 Article IV staff report (IMF Country Report 15/289) indicates on page 5 that the adjustment was a response to a sharp decline in oil prices and other external shocks. Page 9 and Box 2 on page 21 of the report present details the measures and their estimated impacts. Given the motivations for the adjustment described above, we chose to exclude this episode from the database.

16

In 1991 authorities adopted temporary measures to increase revenue, including a tax amnesty and a one-time 2 percent tax on exports with an estimated impact of 1 percent of GDP. Nevertheless, these actions were motivated by economic conditions as described in the Memorandum on Economic and Financial Policies (MEFP) on page 35 of the 1992 Article IV consultation staff report (EBS/92/187): “The Administration of President Serrano, upon assuming office in January 1991, was confronted with a severe economic crisis resulting from persistent large fiscal deficits, poor management of exchange rate policy, and negative real interest rates which led to a decline in domestic savings and encouraged capital flight. … Against this background, the new Administration initiated a stabilization program aimed at reducing inflation and improving the external position”.

Moreover, A tax reform was implemented in July 1992 with an estimated revenue yield of 1.3 percent of GDP in 1993, motivated by the desire to reduce the deficit (see pages 3 and 4 of the 1992 Article IV consultation staff report EBS/92/187), but the impact of the reform was more than offset by increases in expenditures (especially wages), as discussed in the 1993 Article IV consultation staff report SM/94/50.

17

The fiscal year (FY) in Jamaica runs from April to March. The budgetary impact of all measures is expressed in calendar year terms.

18

A comprehensive tax package amounting to about 4 percent of GDP implemented in 1993 was more than offset by significant increases in public sector salaries and was therefore excluded from the database. See the 1993 Article IV staff report (EBS/93/137) page 5 and the 1994 Article IV staff report (EBS/95/6) page 2. Furthermore, in 1998, the Jamaican authorities adopted measures to increase revenues (mainly changes in tax administration) amounting to 3 percent of GDP as well as capital expenditure cuts amounting to 1 percent of GDP. The measures were implemented in the context of a program to reduce inflation, promote growth and maintain high international reserves. See the 1998 Article IV consultation staff report (SM/98/165) pages 17–19 for a discussion. We consider that these measures were motivated endogenous considerations and chose to exclude them from the database (see Section IV for a more detailed description of this episode). In addition, the government introduced tax measures with an expected yield of 3 percent of GDP over the 2009/10 fiscal year, but the tax package seems to have been at least in part motivated by cyclical considerations and was therefore excluded from the database. The 2009 Article IV consultation staff report (IMF Country Report No. 10/267) sheds light on the motivation for the adjustment measures adopted on page 5: “Government finances have deteriorated, constraining the authorities’ ability to respond to the global shock with countercyclical policies. During FY2008/09, the public sector deficit widened by 1¼ percent of GDP, to 9½ percent. During this current fiscal year, faced with declining revenues and a sharp increase in interest costs, the government introduced two packages of fiscal measures equivalent to 1.9 percent of GDP”. According to page 5 of the 2010 Memorandum of Economic and Financial Policies: “To safeguard public finances, the government has introduced three packages of measures over the past year, aimed at boosting revenue by over 3 percent of GDP. In the first phase of measures, introduced at the time of the FY 2009/10 budget, the government raised the excise tax on gasoline by J$8.75 per litre, and broadened the General Consumption Tax base by eliminating exemptions on several items. The second package, aimed at generating annualized revenue equivalent to 0.3 percent of GDP, became effective on October 1, 2009 and included increasing the GCT rate on telephone services from 20 percent to 25 percent and increasing the departure tax to J$1,800. The third package aimed at generating revenue equivalent to 1¾ percent of GDP, became effective on January 1, 2010 (prior action)”. Finally, in July 2016 the first phase of a tax reform aiming at rebalancing the tax structure from direct to more growth-friendly indirect taxes was implemented (IMF Country Report 16/350, page 9). Nevertheless, the reform was designed to be revenue neutral and therefore was not included in the database.

19

An increase in the general VAT tax rate from 10 to 15 percent effective as of May, was one of the most prominent adjustment measures in 1995. The adjustment was undertaken in the context of an IMF-supported program. The estimated impact of this increase in the VAT was of 1.2 percent of GDP (see page 7 of the first 1995 SBA review report, EBS/95/47). Moreover, adjustments to public sector prices were projected to generate revenues of 0.3 percent of GDP. In addition, cuts in primary expenditures in the order of 1.4 percent of GDP were implemented (see Table 3 on page 29 of the 1997 Article IV staff report, SM/97/201), in particular, reductions on the wage bill (0.7 percent of GDP) and capital expenditures (0.6 percent of GDP). But these adjustment measures were driven by current and prospective economic conditions as discussed on page 2 of the 1995 Article IV consultation staff report (EBS/95/103) on page 2. Hence, we chose to exclude the measures from the database (see Section IV for a more detailed description of this episode). Gunter et al. (2017) also flag the VAT hike as endogenous.

Tax measures were also adopted in December 2001 with an expected yield of 1 percent of GDP, but the package was fully allocated to finance additional social expenditure and investment (IMF Country Report 02/237, page 15) and therefore these measures were excluded from the database. Another fiscal reform package was approved in September 2007 with an estimated revenue gain of 1 percent of GDP, but it was also linked to expenditure increases (IMF Country Report 07/379, pages 25 and 27) and therefore excluded.

20

Fiscal adjustment measures were adopted in 1990 in the context of a program to reduce inflation and build-up international reserves. We consider that the adjustment was a direct response to economic conditions and hence we excluded this episode from the database (see Section IV for a more detailed discussion of this episode). The 1991 SBA request report (EBS/91/5) presents on pages 5 and 6 the motivations for the program. The operating surplus of public enterprises increased by 2.1 percent of GDP, whereas tax revenue remained broadly constant as a share of GDP. Moreover, public sector expenditure fell by about 1 percent of GDP (of which about 0.6 percent of GDP were cuts in current expenditure). In addition, A new administration took office in August 1998 and implemented a number of measures to increase revenue for the rest of the year, including increases in excise duties and tax administration improvements. The motivation for policy measures adopted seems to be related to current economic conditions (especially a slow-down in economic activity and trade with Brazil) as discussed on page 9 of the 1999 Recent Economic developments report (IMF country report 99/10). Hence, we chose to exclude this episode from the database (see Section IV for a more detailed discussion of this episode). Moreover, a tax reform introducing a personal income tax (motivated by long-run considerations) was enacted in July 2012, but was accompanied by an increase in current expenditures, in particular wages (see Table 1 on page 29 of the 2013 Article IV staff report, IMF Country Report 14/60). The 2012 Article IV staff report (IMF Country Report 12/211) states on page 14 that: “The PIT was originally approved in 2004 but its implementation has been postponed several times by congress.” The original estimates of the impact of the introduction of the PIT were between 0.2 to 0.3 percent of GDP (IMF Country Report 10/170).

21

Historical data from the Situacion Financiera for the central government downloaded from the Ministry of Finance website (http://www.hacienda.gov.py/situfin/) indicates that interest payments for the central government remained broadly constant as a share of GDP in 1989 relative to 1988.

22

A 1993 tax reform, in the context of a stabilization program supported by the IMF, with an estimated revenue yield of 1.6 percent of GDP, was fully offset by increases in capital and current expenditures (see EBS/94/137, pages 7 and 13 and EBS/95/177 pages 2 and 14) and therefore excluded from the database. In addition, in mid-2014 Peruvian authorities announced several tax cuts leading to an estimated revenue loss of 0.7 percent of GDP in 2015 as well as expenditure increases. These measures were not included in the database because they were driven by short-term stabilization motives. According to the 2015 Selected Issues Paper (IMF Country report 15/134, page 61): “The objective was to spur domestic demand, further streamline regulations, and ease the tax burden in several tax categories”. See also Box 1 in the 2015 Article IV consultation staff report (IMF country report 15/133) and page 6 and Box 8.1 of the 2016–18 Marco Macroeconomico Multinaual report.

23

A tax reform motivated by long-run considerations was implemented in July 2007, but with an overall neutral impact on revenues and therefore it was not included in the database. The reform introduced a personal income tax and broadened the VAT tax base, but reduced corporate income taxes and reduced the VAT rate from 23 to 22 percent. According to the 2007 Article IV report (IMF Country Report No. 08/45), page 16: “… And, while the recently approved tax reform is expected to improve the efficiency of the system, the tax burden is projected to remain largely unchanged.” See also the report for the ex-post evaluation of the SBA (IMF Country Report 08/47) page 19 and the report for the 5th and 6th Reviews of the SBA (EBS/06/166) page 15. Gunter, Riera-Crichton, Vegh, and Vuletin (2017) consider the 1 percentage point reduction in the VAT to be exogenous driven by long-run considerations. While we agree with their assessment of the motivation of the policy actions, we believe that when the complete tax package is taken in to account (beyond the mere decrease in the statutory VAT rate), the policy record does not suggest a significant budgetary impact of the actions. The staff report for the 2008 Article IV consultation on page 29 Table shows 3 shows that revenues from VAT and excise taxes decreased by 0.1 percentage points of GDP in 2007, while overall tax revenue increased by 0.9 percent of GDP. This was accompanied by an increase in primary expenditures of 2.2 percent of GDP (especially wages by 1 percent of GDP).

24

Note that targets were set for 1995/96 in the discussion in the text because authorities had elaborated a program for the period up to March 1996 that was monitored informally by IMF staff.

A New Action-based Dataset of Fiscal Consolidation in Latin America and the Caribbean
Author: Mr. Antonio David and Mr. Daniel Leigh