## 5. Poverty and Inequality in Latin America: Gains during the Commodity Boom but an Uncertain Outlook

## Source details

**Canonical URL:** [5. Poverty and Inequality in Latin America: Gains during the Commodity Boom but an Uncertain Outlook](https://www.imf.org/-/media/files/publications/reo/whd/2018/may/wreo0518-chp5.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/reo/whd/2018/may/wreo0518-chp5.pdf.md)
- [Structured JSON version](/-/media/files/publications/reo/whd/2018/may/wreo0518-chp5.pdf.json)

---

### Panoramic view of social gains during the commodity boom
- Timeframe and definition:
  - Boom period end defined as the start of the 2014 oil price shock; primary analysis period 2000–14 (comparison with the 1990s).
- Geographic and data scope:
  - Focus on Latin America (limited Caribbean data).
  - Country coverage (poverty/inequality analysis): Argentina, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, the Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, and Uruguay.
  - Commodity exporters (net commodity exports > 10 percent of total exports plus imports as of October 2015 WEO) plus Brazil: Argentina, Brazil, Bolivia, Chile, Colombia, Ecuador, Honduras, Paraguay, and Peru.
- Empirical patterns:
  - Poverty reduction was strong across the region during the commodity boom, especially in South America.
  - Inequality (income Gini) declined in both Central and South America, with a significantly larger decline in South America.
  - South America recorded a stark improvement relative to the 1990s when poverty and inequality had increased.

### Growth, commodity terms of trade, and differential social outcomes
- Growth and poverty:
  - During the commodity boom, average real GDP growth increased in South America (where poverty fell the most); Central America growth was lower but remained high.
  - For individual countries, higher GDP growth was positively associated with poverty reduction; South American countries reduced poverty more per percentage point of growth than other emerging markets.
- Commodity terms-of-trade association:
  - South America experienced significant commodity terms-of-trade gains during 2000–14 relative to other regions.
  - Commodity exporters made larger gains in poverty reduction across the board (exceptions: Chile and Honduras).
  - Bolivia and Ecuador—highly dependent on commodity exports—registered the largest gains in poverty and inequality improvements.
  - For inequality, the pattern is mixed: some noncommodity exporters (El Salvador, Dominican Republic) saw larger reductions than several commodity exporters (Chile, Colombia, Paraguay, Honduras).
  - Statistical association is strong for commodity exporters: the size of poverty reduction is directly proportional to the growth rate of the commodity terms of trade; the relationship for inequality is weaker but visible. For noncommodity exporters, no clear association is observed.

### Micro-evidence and income-share regressions (commodity exporters)
- Regression setup:
  - Sample restricted to commodity exporters; dependent variables: income share by decile regressed on (Log) Net commodity Price index, with country fixed effects and GDP per capita control; period 2000–14.
  - Number of countries = 9; Observations = 114 for each decile regression.
- Key coefficient estimates from Table 5.1 (impact of (Log) Net commodity Price index on income share by decile; standard errors in parentheses; significance levels preserved):
  - Decile 1: 0.151 (0.120)
  - Decile 2: 0.395** (0.191)
  - Decile 3: 0.392* (0.207)
  - Decile 4: 0.405* (0.226)
  - Decile 5: 0.476** (0.236)
  - Decile 6: 0.575** (0.255)
  - Decile 7: 0.716*** (0.267)
  - Decile 8: 0.790*** (0.259)
  - Decile 9: 0.436 (0.301)
  - Decile 10: −4.310** (1.735)
- Goodness of fit (R-squared) by decile regressions:
  - Decile 1: 0.608
  - Decile 2: 0.627
  - Decile 3: 0.664
  - Decile 4: 0.674
  - Decile 5: 0.685
  - Decile 6: 0.658
  - Decile 7: 0.604
  - Decile 8: 0.488
  - Decile 9: 0.020
  - Decile 10: 0.638
- Interpretation:
  - Income shares of deciles 2 through 8 increased significantly with higher commodity prices; the share of the top decile (Decile 10) declined substantially on average.
  - The bottom income decile (Decile 1) did not see a statistically significant rise in income share, although its absolute income did increase.
  - Poverty reduction was driven more by income gains closer to the poverty line (deciles 2–4).

### Channels through which commodity cycles affect poverty and inequality
- Market and private sector channels:
  - Expansion of the commodity sector draws labor and resources, raising labor demand, real wages, and/or employment.
  - Improved terms of trade and commodity-sector expansion increase domestic demand, expanding the nontradable sector; commodity-sector investment can raise construction and related activity.
  - Relative wage changes (possible compression of the skills premium if expanding sectors are unskilled-labor intensive) can benefit lower-skilled workers and reduce inequality.
  - Ambiguous effects on the noncommodity tradable sector: possible "Dutch disease" versus positive local input spillovers to manufacturing.
- Fiscal channels:
  - Higher commodity revenues increase government investment, boosting domestic demand and wages via public construction.
  - Larger government transfers—especially if targeted at lower-income groups—directly reduce poverty and inequality.
- Other general equilibrium effects:
  - Migration and financial system propagation can transmit wealth shocks (not examined in depth in this chapter).

### Regional macroeconomic evidence consistent with channels
- Aggregate observations for commodity exporters (2000–14):
  - Public investment and employment growth were higher in commodity exporters than in noncommodity exporters.
  - Commodity exporters experienced significantly larger real labor income gains than noncommodity exporters across all skill levels.
  - Low-skilled workers gained the most in commodity exporters, compressing the skills premium and contributing to inequality reduction.
  - Government transfers increased more in commodity exporters than noncommodity exporters.
- Caveats:
  - Gains in inequality also reflect supply-side factors (e.g., increasing supply of skilled workers) and policy actions (e.g., expansion of cash transfers, policies to boost low wages).

### Micro-data case studies: Bolivia, Brazil, and Peru — channels of social gains
- Countries analyzed: Bolivia, Brazil, Peru (commodity exporters; Brazil more diversified).
- Methodology:
  - Shapley decompositions on household survey data for Bolivia (2013 vs 2007) and Peru (2011 vs 2007); municipal-level analysis for Brazil and Bolivia to compare commodity-producing and non-commodity-producing regions.
- Key findings:
  - For Bolivia and Peru, labor income played a larger role than nonlabor income in reducing inequality and poverty.
  - Changes in labor income of the nontradable (services) sector explain much of the social progress.
  - Low-skilled workers (complete primary or incomplete secondary education) were major contributors to the fall in poverty and inequality.
  - Skilled workers also contributed to poverty reduction because wages at the lower end of their distribution moved up during the boom, enabling some skilled workers to exit poverty.

### Wage, employment, and transfer developments (Bolivia and Peru)
- Bolivia:
  - Real labor income increased for all skill segments except the highest during the boom; largest gains were for intermediate education levels.
  - Sectoral employment growth biggest winners: extractive sector and commerce; in absolute job numbers broad services contributed most.
  - Average wages in the extractive sector fell, likely a compositional effect (informal miners increasing faster than employees in large capital-intensive mines).
  - Transfers increased markedly during the boom, partly reflecting introduction of a noncontributory pension scheme.
- Peru:
  - Sectoral employment growth biggest winners: construction and the extractive sector; broad services created most jobs in absolute terms.
  - Government transfers did not increase substantially during the boom.
- Income composition (selected figures reproduced exactly as presented):
  - Bolivia (2006, 2007, 2011, 2012, 2013): Labor 82.8, 82.4, 81.8, 80.9, 79.1; Nonlabor 16.4, 17.0, 17.9, 18.4, 20.4; Of which: Transfers from government  5.7  5.4  9.8 11.2 . . .
  - Peru (2007, 2008, 2009, 2010, 2011): Labor 83.6, 84.2, 84.9, 84.8, 85.8; Nonlabor 16.4, 15.8, 15.1, 15.2, 14.2; Of which: Current transfers  9.4, 9.0, 9.0, 8.6, 8.3; Of which: Programa JUNTOS  0.5, 0.7, 0.3, 0.3, 0.3
- Implication: transfers account for a much smaller share of income than labor income, mechanically limiting their scope to lower poverty and inequality relative to labor-income channels.

### Shapley decomposition and municipal evidence
- Shapley results (Bolivia and Peru):
  - Labor income changes explain most reductions in poverty and inequality.
  - Nontradable (services) sector labor income changes are major contributors.
  - Low-skilled workers: major contributors to poverty and inequality decline; skilled workers: lower-tail wage growth aided poverty reduction.
- Municipal-level patterns and natural resource producers:
  - Poverty fell in 97 percent of Bolivian municipalities and in 99 percent of Brazilian municipalities between census rounds.
  - Average municipal poverty fall: Bolivia 14 percentage points; Brazil 18 percentage points.
  - In Brazil, out of more than 5,500 municipalities, the top 20 producers account for 75 percent of total production.
  - In Bolivia, the region of Tarija produced about 70 percent of total natural gas in 2012.
- Impact of resource production on municipal outcomes (selected exact estimates):
  - Brazil — Impact of increase in real per capita natural resource production (range for top 20 increases): Poverty: 20.39*** to 29.1***; Gini Coefficient: 0 to 20.05**.
  - Brazil — Being a natural resource producer municipality (dummy): Poverty: 21.44***; Gini Coefficient: 0.
  - Bolivia — Being a natural resource producer municipality (dummy): Poverty: 22.75*.

### Within-country analysis to disentangle fiscal vs labor-demand channels
- Strategy:
  - Distinguish resource types with different expected channels:
    - Brazil: offshore oil and gas (minimal local labor demand; large fiscal windfalls) vs domestic mineral mining (substantial local labor demand).
    - Bolivia: onshore gas megacampos (large fiscal windfalls) vs mineral mining (labor demand).
- Evidence (Brazil municipal regressions, selected coefficients with standard errors):
  - Change in mineral production per capita → Natural Resource Royalties per Capita: 0.0174*** (0.000922)
  - Change in mineral production per capita → Current Revenues per Capita: 0.0241*** (0.006010)
  - Change in mineral production per capita → Share of Workers in Extractive Industries: 1.33e-05*** (0.000004)
  - Change in offshore oil and gas production per capita → Natural Resource Royalties per Capita: 0.0209*** (0.001300)
  - Change in offshore oil and gas production per capita → Current Revenues per Capita: 0.0248*** (0.002640)
  - Change in offshore oil and gas production per capita → Share of Workers in Extractive Industries: 22.56E-06 (0.000002)
  - Model controls: Geographic controls Yes; Dependent variable in 2000 Yes; Change in dependent variable between 1991 and 2000 No; State fixed effects Yes
  - Observations: 5,507; 4,982; 5,507. R-squared: 0.886; 0.834; 0.223.
- Interpretation:
  - Offshore oil and gas production proxies the pure fiscal channel and is associated with municipal revenue increases with limited local labor-demand effects.
  - Mineral mining generates both fiscal revenues and local labor-demand effects.
  - In Brazil, the pure fiscal impact (offshore oil and gas) leads to some reduction in poverty and a marginal increase in labor formality.

### Local labor‑market and poverty effects of natural resource windfalls
- Labor shifted out of agriculture and into nontradables (services and construction) during the commodity boom through both fiscal and market channels, with the market channel especially important.
- Labor formality increased significantly in mineral municipalities.
- Heterogeneity of effects:
  - For most municipalities, a one standard deviation increase in the value of mineral production per capita reduces the poverty rate by 0.2 of a percentage point.
  - For the big producers, the estimated reduction in poverty is between 3 and 9 percentage points for the top five producers.
- Brazil (2000–2010 municipal analysis):
  - Natural resource extraction increases were associated with decreases in poverty, shifts in employment shares away from agriculture and manufacturing into construction and services, increases in labor formality and public sector employment.
- Bolivia:
  - Poverty fell by more in gas megacampo municipalities; labor market impact (decline in agricultural employment share and increase in net migration) is greater in mining municipalities.
  - Public sector employment increased by around 2 percentage points in gas megacampo municipalities, which is greater than one standard deviation.

### Fiscal decentralization, revenue sharing, and local fiscal outcomes
- Redistribution patterns:
  - Bolivia, Brazil, and Peru redistribute large parts of natural resource windfalls back to subnational producers; Colombia redistributes royalties to subnationals but with less focus on producers after a 2012 reform.
- Conceptual drawbacks of large subnational revenue sharing:
  - Geographic and geological differences create large horizontal inequities.
  - Volatility of natural resource revenues complicates intertemporal planning and is harder to manage at local level.
  - Resource revenues to local governments may not encourage building own‑revenue bases or accountability.
  - Large per capita windfalls can lead to absorptive capacity and governance problems.
- Examples and fiscal outcomes (exact figures reproduced):
  - Bolivia departmental budgets (2012): Tarija population share: around 5 percent; its budget accounted for over a third of all departmental revenues and wages, and nearly half of all departmental capital expenditure.
  - Peru (2012): main natural‑resource‑producing departments (Moquegua and Cusco) received more than S/ 2,000 per capita in commodity‑related transfers (canons), while some other departments received less than S/ 1 per capita; 12 of the 183 provinces receive about 50 percent of canon revenues.
  - Some local governments with largest windfalls per capita accumulated large deposits during the boom while other regions had acute investment needs.
  - Since the boom, major commodity-producing regions (Tarija and Rio de Janeiro) have faced severe fiscal sustainability problems.

### Policy implications and recommended reforms for decentralization and revenue management
- When substantive reforms to decentralization frameworks are feasible, they should:
  - Minimize horizontal inequities.
  - Avoid boom‑bust revenue cycles at the local level.
  - Clarify goals of revenue‑sharing agreements.
- Measures to reduce local boom‑bust cycles and governance risks:
  - Use precautionary stabilization funds with clear rules and governance arrangements (examples cited: Chile, Colombia, Norway).
  - Reform royalty‑sharing arrangements to reduce inequities (Colombia’s 2012 reform offered an example).
  - Build capacity at the subnational level and encourage local own‑revenue bases (for example, via property taxes).
  - Make transfer arrangements transparent to facilitate planning and oversight.
  - Consider nonresource transfers to offset horizontal inequities using measurable criteria of local needs (for example, equalization schemes).

### Risks to social progress from lower commodity prices and policy responses
- Post‑boom trends and risks:
  - Since the end of the boom in 2014, commodity prices have been significantly lower.
  - Employment growth slowed much more in commodity exporters than importers.
  - Real wage growth has been negative for all skill groups (see Figures 5.24 and 5.25 in source).
  - The poverty cycle has reversed in some commodity exporters, with increases in poverty rates in Brazil and Paraguay.
  - Inequality in commodity exporters has largely moved sideways post‑2014 after large reductions during the boom.
  - Fiscal space in many commodity exporters has fallen due to lower commodity‑related revenues and slowing growth.
  - Absent policy measures, lower commodity prices pose a significant risk of slower poverty reduction and possibly higher inequality in commodity exporters.
- Policy responses and tools to sustain social progress:
  - Maintain quality of social and infrastructure spending by increasing revenues and reprioritizing spending.
  - Potential revenue measures:
    - Increase revenues from progressive personal income taxes (effective rate for top decile in Latin America averages 5.4 percent according to cited evidence).
    - Reduce universal price subsidies (for example, energy subsidies) which are typically highly regressive.
  - Improve spending efficiency and better target existing social transfers (for example, through means testing where feasible).
  - Fiscal federalism and stabilization:
    - Reform allocation of revenue‑capacity and spending responsibilities across levels of government; enhance subnational capacity.
    - Reform revenue‑sharing formulas to better account for spending needs (population size and poverty levels).
    - Consider greater use of stabilization funds in commodity exporters with clear rules and governance.
  - Labor‑market and structural policies:
    - Increase labor market flexibility and deploy policies for retooling workers to smooth adjustment as demand rebalances.
    - Continue structural reforms to diversify the production base to increase resilience to commodity price shocks.
  - Human capital:
    - Maintain focus on improving the quality of education, recognizing that gains take time and accrue in the longer run.

### Annex: Details of Natural Resource Revenue Sharing (selected country facts)
- Overview:
  - Natural resource revenues are largely centralized in Chile, Ecuador, Mexico, Norway, Trinidad and Tobago, and Venezuela, with limited or no redistribution to subnational producers.
  - In the three case study countries and Colombia, significant amounts go to subnational governments.
  - In Canada, provinces manage nonrenewable natural resources.
- Bolivia (exact allocations):
  - Total hydrocarbon royalty: 18 percent
    - 11 percentage points go to producing departments
    - 6 percentage points stay with the central government
    - 1 percentage point goes to the lightly populated departments of Pando and Beni
  - Hydrocarbon tax (Impuesto directo a los hidrocarburos—IDH): 32 percent
    - Allocated to both producing and nonproducing departments as well as municipalities
    - 20 percentage points remain with the central government
  - Mining royalties:
    - Distributed only to producing departments and municipalities
    - Split between departments and municipalities: 85–15
- Brazil:
  - Mineral royalties distribution:
    - 65 percent distributed directly to the producing municipality
    - 23 percent go to the producing state
    - Remainder to the federal government
  - Oil and gas allocation formula is more complicated; since the 1997 royalties law, substantial amounts have been distributed to municipalities hosting onshore fields or facing offshore fields.
  - In some cases, royalties can account for over 50 percent of a municipality’s revenues.
- Canada:
  - Natural resource income subject to federal and provincial corporate income tax; mining taxes, royalties, and land taxes at the provincial level.
  - Federal fiscal stabilization program: federal assistance if year‑over‑year decline in nonresource revenues > 5 percent caused by an economic downturn.
  - Equalization program: unconditional transfers determined by provinces’ ability to raise revenues.
- Colombia (post‑2012 reform exact outcomes):
  - Prior to the 2012 reform: roughly 80 percent of royalties went directly to producer departments and municipalities (which had 17 percent of the population).
  - Following the 2012 reform: reduced to roughly 10 percent to producer departments and municipalities.
    - Around 30 percent saved in a stabilization fund
    - 10 percent goes to a science and innovation fund
    - 10 percent to a regional pension fund
    - Remainder allocated to subnational investment projects with a complex distribution formula based on poverty levels and other factors
  - Result: 1,089 municipalities received a share of commodity royalties in 2012 compared to 522 in 2011.
- Norway:
  - Government revenues from petroleum activities are transferred to the Government Pension Fund Global.
  - Fiscal rule: petroleum revenues are phased into the economy gradually; government spending must not use any of the fund’s capital, only its expected real return, currently estimated at 3 percent.
- Peru:
  - About 60 percent of fiscal revenues from the mining sector go to subnational governments (mainly mining sector corporate income taxes (canon minero) and mining royalties).
  - Canons transferred only to the department where production takes place; producing provinces and municipalities receive a large share.

### Key statistics and exact figures cited
- Regression sample: Number of countries = 9; Observations = 114 for each decile regression.
- Income‑share regression coefficients (selected): Decile 2: 0.395** (0.191); Decile 7: 0.716*** (0.267); Decile 10: −4.310** (1.735).
- R-squared by decile regressions: Decile 1: 0.608; Decile 9: 0.020; Decile 10: 0.638.
- Municipal impacts and local effects:
  - One standard deviation increase in mineral production per capita → poverty rate fall of 0.2 of a percentage point (typical municipality).
  - Top five producers → estimated reduction in poverty between 3 and 9 percentage points.
  - Increase of around 2 percentage points in public sector employment in gas megacampo municipalities (Bolivia), greater than one standard deviation.
  - Tarija population share: around 5 percent.
  - Peru commodity‑related transfers in 2012: main producing departments received more than S/ 2,000 per capita; some departments received less than S/ 1 per capita; 12 of the 183 provinces receive about 50 percent of canon revenues.
- Post‑boom (post‑2014) developments:
  - Employment growth slowed much more in commodity exporters than importers.
  - Real wage growth has been negative for all skill groups (post‑2014 period).
  - Effective personal income tax revenue for top decile in Latin America averages 5.4 percent (cited evidence).

*Based on a forthcoming IMF Departmental Paper and International Monetary Fund | April 2018 — Chapter: 5. Poverty and Inequality in Latin America: Gains during the Commodity Boom but an Uncertain Outlook.*

### chapter is based on a forthcoming IMF Departmental Paper that

### 5. Poverty and Inequality in Latin America: Gains during the Commodity Boom but an Uncertain Outlook

### Panoramic view of social gains during the commodity boom
- Timeframe and definition:
  - Boom period defined with the end of the boom as the start of the 2014 oil price shock; primary analysis period 2000–14 (comparison with the 1990s).
- Geographic and data scope:
  - Focus on Latin America (limited Caribbean data).
  - Country coverage (poverty/inequality analysis): Argentina, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, the Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, and Uruguay.
  - Commodity exporters (net commodity exports > 10 percent of total exports plus imports as of October 2015 WEO) plus Brazil: Argentina, Brazil, Bolivia, Chile, Colombia, Ecuador, Honduras, Paraguay, and Peru.
- Empirical patterns:
  - Overall, poverty reduction was strong across the region during the commodity boom, especially in South America.
  - Inequality (income Gini) declined in both Central and South America, but the decline was significantly larger in South America.
  - South America recorded a particularly stark improvement relative to the 1990s when poverty and inequality had increased.

### Growth, commodity terms of trade, and differential social outcomes
- Growth and poverty:
  - During the commodity boom, average real GDP growth increased in South America (where poverty fell the most); Central America growth was lower but remained high.
  - For individual countries, higher GDP growth was positively associated with poverty reduction; South American countries reduced poverty more per percentage point of growth than other emerging markets (they lie below the fitted line in the growth–poverty association).
- Commodity terms-of-trade association:
  - South America experienced significant commodity terms-of-trade gains during 2000–14 relative to other regions.
  - Commodity exporters made larger gains in poverty reduction across the board (exceptions: Chile and Honduras). Bolivia and Ecuador—highly dependent on commodity exports—registered the largest gains in poverty and inequality improvements.
  - For inequality, the pattern is mixed: some noncommodity exporters (El Salvador, Dominican Republic) saw larger reductions than several commodity exporters (Chile, Colombia, Paraguay, Honduras).
  - Statistical association is strong for commodity exporters: the size of poverty reduction is directly proportional to the growth rate of the commodity terms of trade; the relationship for inequality is weaker but visible. For noncommodity exporters, no clear association is observed.

### Micro-evidence and income-share regressions (commodity exporters)
- Regression setup:
  - Sample restricted to commodity exporters (because no statistical association for noncommodity exporters).
  - Regression of income share by decile on (Log) Net commodity Price index, with country fixed effects and GDP per capita control; period 2000–14.
  - Number of countries = 9; Observations = 114 for each decile regression.
- Key coefficient estimates from Table 5.1 (impact of (Log) Net commodity Price index on income share by decile; standard errors in parentheses; significance levels preserved):
  - Decile 1: 0.151 (0.120)
  - Decile 2: 0.395** (0.191)
  - Decile 3: 0.392* (0.207)
  - Decile 4: 0.405* (0.226)
  - Decile 5: 0.476** (0.236)
  - Decile 6: 0.575** (0.255)
  - Decile 7: 0.716*** (0.267)
  - Decile 8: 0.790*** (0.259)
  - Decile 9: 0.436 (0.301)
  - Decile 10: −4.310** (1.735)
- Goodness of fit (R-squared) by decile regressions:
  - Decile 1: 0.608
  - Decile 2: 0.627
  - Decile 3: 0.664
  - Decile 4: 0.674
  - Decile 5: 0.685
  - Decile 6: 0.658
  - Decile 7: 0.604
  - Decile 8: 0.488
  - Decile 9: 0.020
  - Decile 10: 0.638
- Interpretation:
  - Income shares of deciles 2 through 8 increased significantly with higher commodity prices; the share of the top decile (Decile 10) declined substantially on average.
  - The bottom income decile (Decile 1) did not see a statistically significant rise in income share, although its absolute income did increase.
  - Poverty reduction was driven more by income gains closer to the poverty line (deciles 2–4).

### Channels through which commodity cycles affect poverty and inequality
- Market and private sector channels:
  - Expansion of the commodity sector draws labor and resources, raising labor demand, real wages, and/or employment.
  - Improved terms of trade and commodity-sector expansion increase domestic demand, expanding the nontradable sector; commodity-sector investment can raise construction and related activity.
  - Relative wage changes (possible compression of the skills premium if expanding sectors are unskilled-labor intensive) can benefit lower-skilled workers and reduce inequality.
  - Ambiguous effects on the noncommodity tradable sector: possible "Dutch disease" (crowding out of tradables via real exchange rate appreciation) versus positive local input spillovers to manufacturing.
- Fiscal channels:
  - Higher commodity revenues increase government investment, which boosts domestic demand and wages via public construction and similar projects.
  - Larger government transfers—especially if targeted at lower-income groups—directly reduce poverty and inequality.
- Other general equilibrium effects:
  - Migration and financial system propagation can transmit wealth shocks (not examined in depth in this chapter).

### Regional macroeconomic evidence consistent with channels
- Aggregate observations for commodity exporters (2000–14):
  - Public investment and employment growth were higher in commodity exporters than in noncommodity exporters.
  - Commodity exporters experienced significantly larger real labor income gains than noncommodity exporters across all skill levels.
  - Low-skilled workers gained the most in commodity exporters, compressing the skills premium and contributing to inequality reduction.
  - Government transfers increased more in commodity exporters than noncommodity exporters.
- Caveats:
  - Gains in inequality also reflect supply-side factors (e.g., increasing supply of skilled workers) and policy actions (e.g., expansion of cash transfers, policies to boost low wages) in various countries and subregions.

### Summary findings
- The commodity boom (2000–14) coincided with large reductions in poverty and meaningful declines in income inequality across Latin America, with the largest social gains in South America and in commodity-exporting countries.
- Statistical analysis for commodity exporters shows that higher commodity terms of trade are associated with increases in income shares for deciles 2–8 and a substantial decline in the income share of the top decile.
- Channels driving these outcomes include labor market effects (employment and wages), fiscal responses (investment and transfers), and general equilibrium adjustments; the poverty effect is stronger and more directly linked to commodity terms-of-trade growth than the inequality effect.

*Based on a forthcoming IMF Departmental Paper that will present further analysis and details on commodity cycles and inequality in Latin America.*

### 5. pOvERTy ANd INEqUALITy IN LATIN AMERICA: GAINs dURING THE COMMOdITy BOOM BUT AN UNCERTAIN OUTLOOK

### 5. pOvERTy ANd INEqUALITy IN LATIN AMERICA: GAINs dURING THE COMMOdITy BOOM BUT AN UNCERTAIN OUTLOOK

### Micro-data case studies: Bolivia, Brazil, and Peru — channels of social gains
- Countries analyzed: Bolivia, Brazil, Peru (commodity exporters; Brazil more diversified).
- Methodology:
  - Shapley decompositions on household survey data for Bolivia (2013 vs 2007) and Peru (2011 vs 2007) to isolate contributions of labor income and nonlabor (transfer) income to changes in poverty and inequality.
  - Municipal-level analysis for Brazil and Bolivia to compare commodity-producing and non-commodity-producing regions.
- Key findings:
  - For both Bolivia and Peru, labor income played a larger role than nonlabor income in reducing inequality and poverty.
  - Across sectors, changes in labor income of the nontradable (services) sector explain much of the social progress.
  - Low-skilled workers (complete primary or incomplete secondary education) were one of the biggest contributors to the fall in poverty and inequality.
  - Skilled workers (complete secondary or tertiary education) also contributed to poverty reduction because wages at the lower end of their distribution moved up during the boom, enabling a nontrivial fraction of skilled workers to exit poverty.

### Wage, employment, and transfer developments (Bolivia and Peru)
- Bolivia:
  - Real labor income increased for all skill segments except the highest during the boom; largest gains were for intermediate education levels.
  - Sectoral employment growth biggest winners: extractive sector and commerce; in absolute job numbers broad services contributed most.
  - Average wages in the extractive sector fell, likely due to a compositional effect (informal miners increasing faster than employees in large capital-intensive mines).
  - Transfers increased markedly during the boom, partly reflecting introduction of a noncontributory pension scheme.
- Peru:
  - Sectoral employment growth biggest winners: construction and the extractive sector; broad services created most jobs in absolute terms.
  - Government transfers did not increase substantially during the boom.
- Income composition (Table 5.2):
  - Bolivia (2006, 2007, 2011, 2012, 2013): Labor 82.8, 82.4, 81.8, 80.9, 79.1; Nonlabor 16.4, 17.0, 17.9, 18.4, 20.4; Of which: Transfers from government  5.7  5.4  9.8 11.2 . . .
  - Peru (2007, 2008, 2009, 2010, 2011): Labor 83.6, 84.2, 84.9, 84.8, 85.8; Nonlabor 16.4, 15.8, 15.1, 15.2, 14.2; Of which: Current transfers  9.4, 9.0, 9.0, 8.6, 8.3; Of which: Programa JUNTOS  0.5, 0.7, 0.3, 0.3, 0.3
- Implication: transfers account for a much smaller share of income than labor income, mechanically limiting their scope to lower poverty and inequality relative to labor-income channels.

### Shapley decomposition results (Bolivia and Peru)
- Labor income changes explain most of the reductions in poverty and inequality.
- Sectoral decomposition: nontradable (services) sector labor income changes are major contributors.
- Skill-level decomposition:
  - Low-skilled workers: major contributors to poverty and inequality decline.
  - Skilled workers: lower-tail wage growth contributed to poverty reduction despite smaller average wage growth.

### Municipal-level patterns and natural resource producers
- Poverty reduction was broad-based across municipalities:
  - Poverty fell in 97 percent of Bolivian municipalities and in 99 percent of Brazilian municipalities between the two census rounds.
  - Average municipal poverty fall: Bolivia 14 percentage points; Brazil 18 percentage points.
- Natural resource production concentration:
  - In Brazil, out of more than 5,500 municipalities, the top 20 producers account for 75 percent of total production.
  - In Bolivia, the region of Tarija produced about 70 percent of total natural gas in 2012.
- Impact of resource production on municipal outcomes (Table 5.3):
  - Impact of increase in real per capita natural resource production (range for top 20 increases):
    - Brazil — Poverty: 20.39*** to 29.1***; Gini Coefficient: 0 to 20.05**; Bolivia — N/A
  - Impact of being a natural resource producer municipality (dummy variable analysis):
    - Brazil — Poverty: 21.44***; Gini Coefficient: 0
    - Bolivia — Poverty: 22.75*
- Summary: social gains in Brazil and Bolivia were broad-based, but natural resource producing municipalities experienced larger declines in poverty.

### Within-country analysis to disentangle fiscal vs labor-demand channels
- Strategy:
  - Distinguish resource types with different expected channels:
    - Brazil: offshore oil and gas (minimal local labor demand; large fiscal windfalls) vs domestic mineral mining (substantial local labor demand).
    - Bolivia: onshore gas megacampos (large fiscal windfalls) vs mineral mining (labor demand).
- Evidence (Table 5.4):
  - Regressions on Brazilian municipalities:
    - Change in mineral production per capita → Natural Resource Royalties per Capita: 0.0174*** (0.000922)
    - Change in mineral production per capita → Current Revenues per Capita: 0.0241*** (0.006010)
    - Change in mineral production per capita → Share of Workers in Extractive Industries: 1.33e-05*** (0.000004)
    - Change in offshore oil and gas production per capita → Natural Resource Royalties per Capita: 0.0209*** (0.001300)
    - Change in offshore oil and gas production per capita → Current Revenues per Capita: 0.0248*** (0.002640)
    - Change in offshore oil and gas production per capita → Share of Workers in Extractive Industries: 22.56E-06 (0.000002)
  - Model controls: Geographic controls Yes; Dependent variable in 2000 Yes; Change in dependent variable between 1991 and 2000 No; State fixed effects Yes
  - Observations: 5,507; 4,982; 5,507. R-squared: 0.886; 0.834; 0.223.
- Interpretation:
  - Offshore oil and gas production proxies the pure fiscal channel and is associated with municipal revenue increases with limited local labor-demand effects.
  - Mineral mining generates both fiscal revenues and local labor-demand effects.
  - In Brazil, the pure fiscal impact (offshore oil and gas) leads to some reduction in poverty and a marginal increase in labor formality.

### Overall synthesis and outlook
- Main drivers of social progress during the boom:
  - Labor income gains, especially for low-to-medium-skilled workers in nontradable (services) sectors.
  - Fiscal windfalls from commodity production contributed to municipal revenues, with differing local labor-demand effects by resource type.
- Limits and concerns:
  - Transfers increased in some cases (Bolivia) but remain smaller than labor income in income composition, limiting their redistributive power relative to labor-market improvements.
  - Sustainability of labor-income gains after the commodity boom is uncertain and is flagged for further analysis in the chapter’s final section.

*International Monetary Fund | April 2018 — Chapter: 5. pOvERTy ANd INEqUALITy IN LATIN AMERICA: GAINs dURING THE COMMOdITy BOOM BUT AN UNCERTAIN OUTLOOK*

### 5. pOvERTy ANd INEqUALITy IN LATIN AMERICA: GAINs dURING THE COMMOdITy BOOM BUT AN UNCERTAIN OUTLOOK

### 5. pOvERTy ANd INEqUALITy IN LATIN AMERICA: GAINs dURING THE COMMOdITy BOOM BUT AN UNCERTAIN OUTLOOK

### Local labor‑market and poverty effects of natural resource windfalls
- Labor shifted out of agriculture and into nontradables (services and construction) during the commodity boom through both fiscal and market channels, with the market channel especially important.
- Labor formality increased significantly in mineral municipalities.
- Effects are heterogeneous:
  - For most municipalities, a one standard deviation increase in the value of mineral production per capita reduces the poverty rate by only 0.2 of a percentage point.
  - For the big producers, the estimated reduction in poverty is between 3 and 9 percentage points for the top five producers.
- In Brazil (2000–2010 municipal analysis):
  - Natural resource extraction increases were associated with decreases in poverty and shifts in employment shares away from agriculture and manufacturing into construction and services, and with increases in labor formality and public sector employment (see Figure 5.21).
- In Bolivia:
  - Poverty fell by more in gas megacampo municipalities, while the labor market impact (decline in fraction of agricultural employment and increase in net migration) is greater in mining municipalities (see Figure 5.22).
  - Public sector employment increased significantly in gas megacampo municipalities; the increase of around 2 percentage points in public sector employment in gas megacampo municipalities is greater than one standard deviation.
- Regression and identification notes:
  - Brazil regressions include change in natural resource production per capita (constant 2010 Brazilian reais), 2000 levels and prior trends (1991–2000), state fixed effects, geographic controls; standard errors clustered at state level.
  - Bolivia analysis uses a difference‑in‑differences model (2001 and 2012 census) with an entropy balancing technique to improve covariate overlap between treatment and control municipalities.

### Fiscal decentralization, revenue sharing, and local fiscal outcomes
- Bolivia, Brazil, and Peru redistribute large parts of natural resource windfalls back to subnational producers; Colombia redistributes royalties to subnationals but with less focus on producers after a 2012 reform.
- Conceptual drawbacks of large subnational revenue sharing:
  - Geographic and geological differences determining fiscal envelopes create large horizontal inequities.
  - Volatility of natural resource revenues complicates intertemporal planning and is harder to manage at local level.
  - Resource revenues to local governments act as transfers and do not encourage building own‑revenue bases or accountability.
  - Large per capita windfalls can lead to absorptive capacity and governance problems.
- Examples and fiscal outcomes:
  - Bolivia departmental budgets (2012): the main gas region (Tarija) has a population share of around 5 percent but its budget accounted for over a third of all departmental revenues and wages, and nearly half of all departmental capital expenditure (see Figure 5.23).
  - Peru (2012): main natural‑resource‑producing departments (Moquegua and Cusco) received more than S/ 2,000 per capita in commodity‑related transfers (canons), while some other departments received less than S/ 1 per capita; 12 of the 183 provinces receive about 50 percent of canon revenues.
  - Some local governments with largest windfalls per capita accumulated large deposits during the boom while other regions had acute investment needs.
  - Since the boom, major commodity-producing regions (Tarija and Rio de Janeiro) have faced severe fiscal sustainability problems.

### Policy implications and recommended reforms for decentralization and revenue management
- When substantive reforms to decentralization frameworks are feasible, they should:
  - Minimize horizontal inequities.
  - Avoid boom‑bust revenue cycles at the local level.
  - Clarify goals of revenue‑sharing agreements.
- Measures to reduce local boom‑bust cycles and governance risks:
  - Use precautionary stabilization funds with clear rules and governance arrangements (examples cited: Chile, Colombia, Norway).
  - Reform royalty‑sharing arrangements to reduce inequities (Colombia’s 2012 reform offered an example).
  - Build capacity at the subnational level and encourage local own‑revenue bases (for example, via property taxes).
  - Make transfer arrangements transparent to facilitate planning and oversight.
  - Consider nonresource transfers to offset horizontal inequities using measurable criteria of local needs (for example, equalization schemes).

### Risks to social progress from lower commodity prices and policy responses
- Summary of trends:
  - Latin America made large gains in reducing inequality and poverty in the 2000s, especially in commodity‑exporting countries, driven by falling labor income inequality (declining skills premium) and expansion of services and lower‑skill jobs; social transfers also contributed.
  - Since the end of the boom in 2014, commodity prices have been significantly lower.
  - Post‑2014 developments:
    - Employment growth slowed much more in commodity exporters than importers.
    - Real wage growth has been negative for all skill groups (see Figures 5.24 and 5.25).
    - The poverty cycle has reversed in some commodity exporters, with increases in poverty rates in Brazil and Paraguay.
    - Inequality in commodity exporters has largely moved sideways post‑2014 after large reductions during the boom.
    - Fiscal space in many commodity exporters has fallen due to lower commodity‑related revenues and slowing growth.
  - Absent policy measures, lower commodity prices pose a significant risk of slower poverty reduction and possibly higher inequality in commodity exporters.

- Policy responses and tools to sustain social progress:
  - Central government actions (especially where fiscal buffers are limited):
    - Maintain quality of social and infrastructure spending by increasing revenues and reprioritizing spending.
    - Potential revenue measures:
      - Increase revenues from progressive personal income taxes (Latin America tends to have lower personal income tax revenue compared to other regions; effective rate for top decile in Latin America averages 5.4 percent according to cited evidence).
      - Reduce universal price subsidies (for example, energy subsidies) which are typically highly regressive.
    - Improve spending efficiency and better target existing social transfers (for example, through means testing where feasible).
  - Fiscal federalism and stabilization:
    - Reform allocation of revenue‑capacity and spending responsibilities across levels of government; enhance subnational capacity.
    - Reform revenue‑sharing formulas to better account for spending needs (population size and poverty levels).
    - Consider greater use of stabilization funds in commodity exporters with clear rules and governance.
  - Labor‑market and structural policies:
    - Increase labor market flexibility and deploy policies for retooling workers to smooth adjustment as demand rebalances.
    - Continue structural reforms to diversify the production base to increase resilience to commodity price shocks.
  - Human capital:
    - Maintain focus on improving the quality of education, recognizing that gains take time and accrue in the longer run.

### Key statistics and exact figures cited
- One standard deviation increase in mineral production per capita → poverty rate fall of 0.2 of a percentage point (typical municipality).
- Top five producers → estimated reduction in poverty between 3 and 9 percentage points.
- Increase of around 2 percentage points in public sector employment in gas megacampo municipalities (Bolivia), which is greater than one standard deviation.
- Tarija population share: around 5 percent.
- Peru commodity‑related transfers in 2012: main producing departments received more than S/ 2,000 per capita; some departments received less than S/ 1 per capita; 12 of 183 provinces receive about 50 percent of canon revenues.
- Post‑boom (2015–16) real labor income growth by educational level: negative across skill groups (see Figure 5.25).
- Employment growth comparisons:
  - 2000–14 vs 2015–16 show commodity exporters experiencing much slower employment growth post‑boom relative to earlier period and relative to noncommodity exporters (see Figure 5.24).

*International Monetary Fund | April 2018*

### Annex 5.2. Details of Natural

### Annex 5.2. Details of Natural Resource Revenue Sharing in Latin America and Elsewhere

### Overview
- Natural resource revenues are largely centralized in Chile, Ecuador, Mexico, Norway, Trinidad and Tobago, and Venezuela, with either very limited or no redistribution to subnational producers.
- In the three case study countries and Colombia, significant amounts go to subnational governments.
- In Canada, provinces manage nonrenewable natural resources.

### Bolivia
- Total hydrocarbon royalty: 18 percent
  - 11 percentage points go to producing departments
  - 6 percentage points stay with the central government
  - 1 percentage point goes to the lightly populated departments of Pando and Beni
- Hydrocarbon tax (Impuesto directo a los hidrocarburos—IDH): 32 percent
  - Allocated to both producing and nonproducing departments as well as municipalities
  - 20 percentage points remain with the central government
- Mining royalties:
  - Distributed only to producing departments and municipalities
  - Split between departments and municipalities: 85–15

### Brazil
- Mineral royalties distribution:
  - 65 percent distributed directly to the producing municipality
  - 23 percent go to the producing state
  - Remainder to the federal government
- Oil and gas allocation formula is more complicated; since the 1997 royalties law, substantial amounts of oil and gas revenues have been distributed to municipalities that either host an onshore oil and gas field or face an offshore oil and gas field.
- In some cases, royalties can account for over 50 percent of a municipality’s revenues.

### Canada
- Natural resource income is subject to:
  - Federal and provincial corporate income tax
  - Mining taxes, royalties, and land taxes at the provincial level
- Fiscal stabilization program:
  - Federal government can provide financial assistance to any province faced with a year-over-year decline in nonresource revenues greater than 5 percent caused by an economic downturn
- Equalization program:
  - Transfers are unconditional and determined by measuring provinces’ ability to raise revenues

### Colombia
- Prior to the 2012 reform:
  - Roughly 80 percent of royalties went directly to producer departments and municipalities, which only had 17 percent of the population
- Following the 2012 reform:
  - Reduced to roughly 10 percent to producer departments and municipalities
  - Remainder assigned to a number of central funds with specific goals
    - Around 30 percent saved in a stabilization fund
    - 10 percent goes to a science and innovation fund
    - 10 percent to a regional pension fund
    - Remainder allocated to subnational investment projects with a relatively complex distribution formula based on poverty levels and other factors
  - Result: 1,089 municipalities received a share of commodity royalties in 2012 compared to 522 in 2011

### Norway
- Government revenues from petroleum activities are transferred to the Government Pension Fund Global.
- Fiscal rule:
  - Petroleum revenues are phased into the economy gradually
  - Over time government spending must not use any of the fund’s capital, only its expected real return, which is currently estimated at 3 percent
  - Rule provides for petroleum revenue spending to be increased during economic downturns and decreased during economic upturns

### Peru
- Overall, about 60 percent of fiscal revenues from the mining sector go to subnational governments
  - Mainly consisting of mining sector corporate income taxes (canon minero) and mining royalties
- Canons:
  - Various canons transferred only to the department where production of the natural resource takes place
  - Resources further distributed within producing departments, resulting in producing provinces and municipalities receiving a large share

*International Monetary Fund | Annex 5.2. Details of Natural Resource Revenue Sharing in Latin America and Elsewhere*

---


_Source: https://www.imf.org/-/media/files/publications/reo/whd/2018/may/wreo0518-chp5.pdf_