## Investment by Economic Activity in Latin America: Closing the Data Gaps

## Source details

**Canonical URL:** [Investment by Economic Activity in Latin America: Closing the Data Gaps](https://www.imf.org/-/media/files/publications/reo/whd/2024/october/english/onlineannex4.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/reo/whd/2024/october/english/onlineannex4.pdf.md)
- [Structured JSON version](/-/media/files/publications/reo/whd/2024/october/english/onlineannex4.pdf.json)

---

### Motivation and objective
- Timely and granular public data on investment by economic activity in Latin America is largely unavailable, hampering assessment of long-term growth potential amid the green transition and nearshoring opportunities.
- Objective: construct quarterly investment data by economic activity for LA5 countries (Brazil, Chile, Colombia, Mexico, Peru) using available macroeconomic statistics.

### Data sources and methodology
- Primary inputs:
  - National accounts (“supply and use” tables) with detailed data on investment products (e.g., buildings, machinery, infrastructure) and origin (domestic or imported).
  - Sector-specific annual investment data from the KLEMS initiative (available with long lags in some cases).
- Construction of annual investment matrices:
  - Combine detailed product-level investment data with KLEMS sectoral investment data to assign products to sectors (examples given: tractors → agriculture; mining equipment → mining; bridges → public administration).
  - Products used by multiple sectors (e.g., computers, furniture) are assigned according to their total sectoral investment.
  - For products serving both investment and consumption purposes (e.g., pickup trucks), supplement with country-specific or additional information to delineate use.
- Temporal and scaling methods:
  - Quarterly estimates generated using quarterly data on domestic production and imports by product groups.
  - Sectoral estimates are calibrated using multivariate temporal disaggregation with constraints to align with total investment figures at the lowest available level of aggregation.
- Output:
  - Quarterly investment data for 10–13 sectors (depending on country) from 2014 onwards.
  - Both nominal and real series can be produced.

### Key findings and patterns
- Broad decline and sectoral composition:
  - Since the end of the commodity price boom a decade ago, the decline in total investment as a share of GDP in LA5 countries (especially when excluding Mexico) was driven by falling investment in non-tradable sectors.
  - The share of tradable sectors remained largely unchanged and even declined for mining, suggesting limited progress in positioning the region to harness growing global demand for critical green minerals.
- Post-pandemic dynamics:
  - Investment recovered relatively strongly after the pandemic, but weakened since 2023, except in Mexico, reflecting weaker commodity prices.
- Country-specific patterns:
  - Brazil, Chile, Peru: pre-pandemic broad-based growth in private investment slowed significantly in recent quarters; notable changes in composition toward (residential) construction in Brazil.
  - Colombia: public sector investment was a key driver of investment growth before and after the pandemic, but its contribution turned negative since mid-2023.
  - Mexico: unlike other LA5 countries, pre-pandemic investment growth was close to zero; it surged since 2021 and remained exceptionally strong, driven by manufacturing (owing to nearshoring) and more recently also by the public sector to complete big infrastructure projects (examples cited: new refineries and railway lines).

### Quantitative indicators and figure values
- Values appearing in Online Annex Figure 4.1 (as presented in the source):
  - 0.9 1.1 1.6 1.5 3.2 3.2 1.3 1.0 5.4 4.6 4.9 4.4 3.2 2.8 20.5 18.5
- Figures and charts produced/used:
  - Online Annex Figure 4.1: LA5 excluding Mexico — Gross Fixed Capital Formation by Economic Activity (Percent of GDP).
  - Online Annex Figure 4.2: Gross Fixed Capital Formation in Mining and Manufacturing (Percent of GDP).
  - Online Annex Figure 4.3: Contributions to Real Gross Fixed Capital Formation Growth by Economic Activity (Year-over-year percent change; 4-quarter moving average) — panels for LA5, Brazil, Chile, Colombia, Mexico, Peru.
- Notes on figures:
  - Aggregates are purchasing-power-parity GDP-weighted averages.
  - “Construction” activity includes residential construction which in original data is included in “services” (real estate) activity.
  - For Colombia, Mexico, and Peru, "social community and personal services" activity is used as a proxy for "public administration."
  - Data for COVID years (2020–21) are not shown in the charts.
  - LA5 = Latin America 5 (Brazil, Chile, Colombia, Mexico, Peru).
  - A footnote notes that one Peru series is based on national accounts compiled by INEI which differs from BCRP data.

### Policy implications and recommendations
- Strengthen national statistics:
  - National authorities should invest in producing more timely and comprehensive statistics on investment by economic activity, with dedicated resources and access to source data.
  - Enhancing official statistics to deliver quarterly, sectoral investment series would improve macroeconomic analysis and policy making.
- Targeted investment priorities:
  - To harness global transformations (green transition, nearshoring), policy should focus on reversing the decline in investment in tradable sectors—notably mining and manufacturing—while addressing the recent shift toward non-tradable and residential construction in some countries.
- Public investment management:
  - Given Colombia’s experience and Mexico’s recent public-driven infrastructure expansion, careful assessment of public investment composition and efficiency is important to sustain productive investment gains.

*Prepared jointly with Lorena Corso and Bright Richard Kimuli under the supervision of Michael Stanger (Statistics Department, IMF); data sources: national authorities; and IMF staff estimates.*

---


_Source: https://www.imf.org/-/media/files/publications/reo/whd/2024/october/english/onlineannex4.pdf_