## CHAPTER 3

## Source details

**Canonical URL:** [CHAPTER 3](https://www.imf.org/-/media/files/publications/weo/2019/april/english/ch3.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/weo/2019/april/english/ch3.pdf.md)
- [Structured JSON version](/-/media/files/publications/weo/2019/april/english/ch3.pdf.json)

---

### Overview and key questions
- The investment needs of most emerging market and developing economies remain substantial; these economies still have only a small fraction of the capital available in advanced economies even though their investment rates have increased significantly over the past three decades, with a near doubling of real investment rates in machinery and equipment.
- Capital deepening in emerging market and developing economies coincided with sizable declines in the price of investment goods, particularly tradable capital goods such as machinery and equipment.
- Key questions examined:
  - How have prices of investment goods evolved over time and across countries? Do lower-income countries face higher capital goods prices, in absolute terms and/or relative to other prices in the economy?
  - What drives the price of tradable capital goods over time, and which factors explain differences across countries? How much have technological advances and trade integration contributed to the relative decline in the prices of machinery and equipment? To what extent are capital goods prices shaped by policy choices, particularly barriers to trade?
  - How responsive is investment in machinery and equipment to the price of these assets? How much have changes in capital goods prices contributed to capital deepening over the past three decades?

### Main findings
- The relative price of tradable investment goods (machinery and equipment) has declined across advanced, emerging market, and developing economies over the past three decades; declines were driven by faster productivity growth in capital goods production and deepening trade integration.
- ICP 2011 data: the same basket of machinery and equipment costs about 8 percent more in the median low-income country than in the median advanced economy.
- Trade costs and lower relative tradable productivity explain a large share of cross-country variation in relative machinery and equipment prices; trade costs and relative tradable productivity can together explain up to 60 percent of the cross-country variation in the 2011 ICP relative price (depending on the trade-cost measure used).
- A 1 percent decline in the relative price of investment goods is associated with:
  - roughly a 0.8 percent increase in the ratio of real investment to real GDP in the medium term in GIMF simulations;
  - a 0.3–0.5 percent increase in the real investment rate over a five-year period in cross-country regressions;
  - a 0.2–0.5 percent increase in the real investment rate in sectoral (EU and World KLEMS) analyses.

### Patterns over time and across asset types
- Four main asset types evolved differently: structures; machinery and equipment (excluding transportation); transportation equipment; intellectual property products. Machinery and equipment and transportation equipment prices declined significantly relative to the consumption deflator since the early 1990s.
- Within tradable tangible capital goods:
  - computing equipment prices fell by 90 percent since 1990;
  - communications equipment prices fell by almost 60 percent.
- By 2014, relative to the early 1990s, the price of machinery and equipment declined relative to the consumption deflator in:
  - all advanced economies;
  - 87 percent of emerging market economies;
  - 68 percent of low-income developing countries.

### Cross-country empirical findings on absolute and relative prices
- Definitions (ICP 2011):
  - Absolute price: price level of machinery and equipment relative to its US level.
  - Relative price: price of machinery and equipment relative to the price of consumption.
- Regression relationships reported in Figure 3.3:
  - Absolute price: y = –0.025***x + 1.46
  - Relative price: y = –0.54***x + 7.30
- ICP 2011: relative to the price of consumption, prices of machinery and equipment are significantly higher in emerging market and developing economies than in advanced economies; lower-income countries face marginally higher absolute prices.
- Comparable cross-country data on capital goods prices are scarce; ICP rounds (1985, 1996, 2011) are the key source.

### Drivers of higher relative investment prices across countries
- Little systematic correlation between exporter unit values and per capita income of importing country for major capital goods exporters (United States, China, Germany, France, Japan).
- Emerging markets and low-income developing countries face systematically higher policy-related and natural trade barriers:
  - policy-related: higher tariffs, documentary and border compliance costs;
  - natural: greater distance to exporters, weaker liner shipping connectivity.
- Trade-cost measures highlighted (median and interquartile ranges by country group) include:
  - Distance to machinery and equipment exporters (Thousand kilometers)
  - Liner shipping connectivity (Index)
  - Freedom to trade internationally (Index)
  - Tariffs on capital goods (Percent)
  - Costs to import (Thousands of US dollars a container)
  - Time to import (Days)
- Trade costs are more powerful predictors of relative prices than natural barriers.
- Export dynamics:
  - Emerging market and developing economies accounted for about one-third of global exports of capital goods in 2016, up from 5 percent in 1990.
  - China’s share rose from 1–2 percent in the 1990s to 18 percent in 2017.

### Over-time drivers of the large decline in relative prices (empirical decomposition)
- Two-step approach (40 advanced and emerging market economies, 1995–2011):
  1. Estimate elasticity of sectoral producer prices to changes in sectoral labor productivity and import penetration (imports/domestic value added), controlling for country-year and country-sector fixed effects.
  2. Combine elasticities with observed changes in relative labor productivity and trade exposure of capital goods sector to quantify contributions to decline in relative prices (2000–11).
- Instrumental-variable strategy: import tariffs used as instrument for import penetration.
- Econometric results (sectoral regressions; all include country-year and country-sector fixed effects; standard errors in parentheses; Number of Observations = 16,077; Reported R^2: 0.62; 0.62; 0.56; 0.91):
  - Relative Import Penetration t – 1:
    - Column (1) OLS: –0.135*** (0.033)
    - Column (2) OLS: –0.107*** (0.037)
    - Column (3) IV: –0.574*** (0.163)
    - Column (4) IV (Relative Productivity): 1.363*** (0.363)
  - Difference for Capital Goods Sectors:
    - Column (1): –0.191** (0.081)
    - Column (2): 0.033 (0.322)
    - Column (4): 1.407** (0.671)
  - Relative Productivity t – 1: –0.316*** (0.035); –0.314*** (0.035); –0.328*** (0.032)
  - Relative Import Penetration for Capital Goods Sectors (additional coefficients): –0.298*** (0.071); –0.541* (0.287); 2.770*** (0.564)
- Decomposition of the 2000–2011 decline in relative price of machinery and equipment into:
  1. Direct effect of deepening trade integration.
  2. Effect of trade integration through higher labor productivity.
  3. Effect of higher labor productivity not due to trade integration.
  4. Residual (other factors).
- Findings from the decomposition:
  - Deepening trade integration accounts for the bulk of the decline, both directly and indirectly via higher labor productivity in domestic capital goods producers.
  - Productivity gains in capital-goods-producing sectors not linked to trade integration are also significant.
  - A nontrivial portion of the decline is attributable to other factors (for example, the downward trend in world interest rates, financial liberalization, and the emergence of China as a key supplier of tradable investment goods).

### Trade, productivity, and sectoral producer prices — key elasticities
- Econometric confirmation (Online Annex 3.5 and Table 3.1):
  - A 1 percent increase in the import ratio (achievable by a 0.7 percentage point cut in tariffs) reduces the sectoral producer price by about 0.5 percent.
  - A 1 percent increase in sectoral labor productivity reduces producer prices by about 0.3 percent.
- Policy-induced increases in import penetration have a strong positive effect on sectoral labor productivity; capital-goods-producing sector labor productivity is particularly sensitive to deepening trade integration.
- Illustrative policy magnitude:
  - If low-income countries brought capital goods tariffs to the level of those in advanced economies (reduce tariffs by roughly 8 percentage points), the price of investment goods would decline by about 16 percent.
    - Roughly 40 percent of that decline would come from the direct trade integration effect and the rest from higher productivity in the capital goods sector due to greater import competition.

### Macroeconomic model simulations (GIMF) — effects on investment and output
- Structural model: IMF’s Global Integrated Monetary and Fiscal model (GIMF).
- Model experiments:
  - Shocks to relative productivity of investment-goods-producing sector and tariff cuts that permanently lower price of capital goods imports lead to sizable and long-lasting increases in real investment rate in a representative emerging market economy.
  - A 1 percent decline in the price of investment relative to consumption leads to a roughly 0.8 percent increase in the ratio of real investment to real GDP in the medium term.
  - For an average emerging market and developing economy with a ratio of real investment to real output of about 22 percent, a 1 percent decline in the relative price of investment would lead to an increase in the investment rate to 22.2 percent.
- Box 3.3 scenarios (medium-term, 10 years after the shock):
  - Scenario 1: investment-specific technological change (higher domestic efficiency in producing new capital).
  - Scenario 2: permanent decline in tariffs on imported capital goods (a 4 percentage point permanent decline in investment goods tariffs required to obtain 1 percent long-term decline in relative price given assumed import share of about 33 percent; this tariff cut entails a recurrent fiscal cost of about 0.25 percent of annual GDP).
  - Simulation outcomes:
    - Both scenarios increase returns to capital and trigger similar increases in investment.
    - Technology shock: real output increases by 0.7 percent of GDP.
    - Capital goods tariff reduction: real output increases by 0.5 percent of GDP.
  - Investment-specific tariff cut vs. equivalent-fiscal-revenue general tariff cut:
    - Investment-specific tariff scenario: Investment increase: 1.34 percent; Output increase: 0.5 percent.
    - General tariff decrease: Investment increase: 0.23 percent; Output increase: 0.18 percent.
  - Interpretation: lowering barriers that specifically reduce the relative price of investment has materially larger investment and output effects than broad tariff reductions that do not lower the domestic relative price of investment.

### Empirical cross-country and sectoral evidence on investment responsiveness
- Cross-country analysis:
  - Data: Over 60 years across 180 economies from Penn World Table.
  - Main empirical result (Table 3.2 and robustness checks):
    - A 1 percent decline in the relative prices of tradable capital goods is associated with a 0.3–0.5 percent increase in the real investment rate over a five-year period.
    - Main IV coefficients (log real investment-to-GDP ratio regressions; IV with lagged price instrument; standard errors in parentheses):
      - Column (1) All: Log Relative Price coefficient = –0.377*** (0.116); Observations = 658; Countries = 127; R^2 = 0.41; First Stage F-Statistic = 118.80
      - Column (2) Post-1990: –0.292* (0.171); Observations = 553; Countries = 127; R^2 = 0.36; First Stage F-Statistic = 81.81
      - Column (3) EMDEs: –0.491*** (0.161); Observations = 457; Countries = 93; R^2 = 0.38; First Stage F-Statistic = 64.04
    - Caveat: empirical estimates likely represent an upper bound because relative investment prices are endogenous.
- Sectoral analysis:
  - Data: EU and World KLEMS covering 18 economies over 1971–2015, 15 broad sectors.
  - Baseline result: A 1 percent decline in the relative price of machinery and equipment is associated with a 0.2–0.5 percent increase in the real investment rate in these capital goods sectors.
  - Declining investment prices linked to higher sectoral output and marginally higher labor productivity.
- Firm-level evidence (Colombia tariff reform 2011):
  - Tariff reform reduced average tariff rate from 12.5 percent in 2010 to 8.7 percent in 2011 (close to a 4 percentage point decline).
  - Event-study estimates:
    - A 1 percentage point reduction in capital goods input tariffs is associated with a 0.4 percentage point increase in investment (statistically significant at the 1 percent level).
    - A 1 percentage point reduction in noncapital input tariffs is associated with a 0.15 percentage point increase in investment (statistically significant).
    - Reductions in output tariffs are not associated with significant short-term changes in firm investment.
  - Interpretation: firm-level evidence shows investment choices are sensitive to capital goods prices; coefficients can be interpreted as price effects under assumption of full tariff pass-through.

### Quantitative contribution of price declines to historical investment growth
- Between 1990–94 and 2010–14:
  - Real investment-to-real GDP ratios in machinery and equipment grew by approximately 60 percent.
  - A significant portion of this increase can be explained by the decline in the relative price of machinery and equipment.
  - Improvements in policies and policy frameworks also contributed significantly.
- Decomposition framework partitions investment growth into:
  1. Decline in real investment prices.
  2. Change in relevant policies.
  3. Other factors (global trends, convergence, growth expectations).
  4. Residual.

### Drivers and threats to continued declines in capital goods prices
- Drivers:
  - Faster productivity growth in capital-goods-producing sectors (linked to information technology advances).
  - Efficiency gains from globalization and specialization; production of machinery and equipment is strongly embedded in global value chains.
- Threats:
  - Slowdown in global trade and maturation of global value chains; waning pace of trade liberalization since the mid-2000s may limit further declines.
  - Hikes in tariffs and nontariff barriers could disrupt cross-border supply chains and slow or reverse downward trend in capital goods prices, harming net importers among emerging market and developing economies.
  - Sluggish productivity growth in advanced economies, lasting scars on R&D and technology adoption, aging populations, and rising market power in main capital-goods-producing economies could weaken innovation needed to sustain price declines.
  - The pace of decline in the relative price of machinery and equipment has slowed considerably in advanced economies in the past decade.

### Policy implications and recommendations
- Reduce trade costs and reinvigorate international trade to maintain declines in relative capital goods prices and support investment and productivity:
  - Many emerging market and developing economies maintain tariff and other trade barriers that raise relative price of capital goods; removing these barriers would encourage investment in tradable capital goods and support capital deepening.
  - Advanced economies should avoid protectionist measures that raise trade costs, since their real investment has also been supported by declining capital goods prices.
  - Reviving trade liberalization (which slowed since the mid-2000s) is vital for maintaining pace of decline in relative capital goods prices.
- Specific policy levers:
  - Lower tariffs and other trade barriers on capital goods (illustration: effective import tariffs on capital goods in 2011 were about 4 percent in emerging market and 8 percent in low-income developing countries, compared with close to zero in advanced economies).
  - Improve logistics and shipping connectivity and implement the World Trade Organization’s Trade Facilitation Agreement (fully implementing commitments could reduce non-tariff barriers by an equivalent of a 15-percentage point tariff cut in less-developed economies).
  - Support research and development, entrepreneurship, technology transfer, education, and public infrastructure investment to boost productivity in capital-goods-producing sectors.
  - Address distributional consequences and job disruptions from declining relative investment prices through long-term investment in education, lifelong skill upgrading, and policies facilitating reallocation of displaced workers to new jobs.
- Fiscal considerations:
  - Tariff cuts targeted at capital goods can entail fiscal costs (example: a 4 percentage point permanent decline in investment goods tariffs implies a recurrent fiscal cost of about 0.25 percent of annual GDP in the model); policymakers should weigh fiscal trade-offs against investment and output gains.

### Additional sectoral notes (Boxes)
- Box 3.1: Between 2009 and 2017, prices of solar photovoltaics fell by 76 percent; prices of onshore wind turbines fell by 34 percent. Divergent cost paths noted across low-carbon technologies.
- Box 3.2: Billion Prices Project (Apple products, 2009–2012) confirms absolute online prices differ across countries with no clear correlation with per capita income, but relative prices of capital goods tend to be significantly lower in richer countries.

*Source: Chapter 3, “The Price of Capital Goods: A Driver of Investment Under Threat?” — World Economic Outlook, April 2019.*

### Introduction

### ch3 - Introduction

### Overview
- The investment needs of most emerging market and developing economies remain substantial. These economies still have only a small fraction of the capital available in advanced economies, even though their investment rates have increased significantly over the past three decades, with a near doubling of real investment rates in machinery and equipment.
- High investment rates have been a key reason for significantly higher growth in emerging market and developing economies than in advanced economies since the early 2000s, underpinning projections that they will grow faster than advanced economies in the medium term.
- The capital deepening in emerging market and developing economies over the past three decades has coincided with sizable declines in the price of investment goods and, in particular, of tradable capital goods, such as machinery and equipment, relative to other prices in the economy.

### Authors and contributors
- The authors of this chapter are Weicheng Lian, Natalija Novta, Evgenia Pugacheva, Yannick Timmer, and Petia Topalova (lead), with support from Jilun Xing and Candice Zhao, and contributions from Michal Andrle, Christian Bogmans, Lama Kiyasseh, Sergii Meleshchuk, and Rafael Portillo. The chapter benefited from comments and suggestions by Andrei Levchenko and Maurice Obstfeld.

### Key questions examined
- How have prices of investment goods evolved over time and across countries? Do lower-income countries face higher capital goods prices, in absolute terms and/or relative to other prices in the economy?
- What drives the price of tradable capital goods over time, and which factors explain differences across countries? How much have technological advances and trade integration contributed to the relative decline in the prices of machinery and equipment? To what extent are capital goods prices shaped by policy choices, particularly barriers to trade?
- How responsive is investment in machinery and equipment to the price of these assets? How much have changes in capital goods prices contributed to capital deepening over the past three decades?

### Main findings
- The relative price of tradable investment goods, namely machinery and equipment, has declined across advanced, emerging market, and developing economies over the past three decades. The declines have been significant and have been driven by faster productivity growth in capital goods production and deepening trade integration.
- According to the International Comparison Program (ICP) data for 2011, emerging market and developing economies faced higher machinery and equipment prices, both in absolute terms and especially relative to the price of consumption. The same basket of machinery and equipment costs about 8 percent more in the median low-income country than in the median advanced economy.
- The higher relative prices of machinery and equipment in these economies reflect lower relative efficiency in producing investment goods and tradable goods more broadly, and significantly higher trade costs, such as those arising from higher tariffs.
- Model simulations and empirical evidence suggest that the relative price of investment goods is an important driver of real investment rates. Over the past 30 years, real investment in machinery and equipment as a share of real GDP has increased significantly in both advanced and emerging market and developing economies. A nontrivial share of this increase can be attributed to the decline in the relative prices of machinery and equipment.

### Drivers and threats to declining capital goods prices
- Drivers:
  - Faster productivity growth in sectors that produce capital goods than in sectors producing consumption goods and services, linked in part to advances in information technology.
  - Efficiency gains from globalization and the associated specialization of production, with production of machinery and equipment strongly embedded in global value chains.
- Threats:
  - The slowdown in global trade, potential maturation of global value chains, and the waning pace of trade liberalization since the mid-2000s may limit further declines in the price of investment.
  - Hikes in tariffs and nontariff barriers could disrupt cross-border supply chains and slow or even reverse the downward trend in capital goods prices, harming net importers of capital goods among emerging market and developing economies.
  - Sluggish productivity growth in advanced economies, lasting scars from the global financial crisis on research and development spending and technology adoption, aging, and the rise of market power in main capital-goods-producing economies could weaken the innovation needed to sustain declines in investment prices.
  - The pace of decline in the relative price of machinery and equipment has slowed considerably in advanced economies in the past decade, potentially exerting an additional drag on investment.

### Policy implications and recommendations
- The findings provide an argument in support of policies aimed at reducing trade costs and reinvigorating international trade. Many emerging market and developing economies still maintain trade barriers that raise the relative price of capital goods for domestic investors; removing these barriers would encourage investment in tradable capital goods and support needed capital deepening.
- Advanced economies should guard against protectionist measures that raise trade costs, since their real investment has also been supported by declining prices of capital goods.
- Reviving the process of trade liberalization, which has slowed down significantly since the mid-2000s, is vital for maintaining the pace of decline in relative capital goods prices, providing impetus to real investment on top of welfare and productivity gains from international trade.
- Continued technological advances and innovation in capital goods production are important in both advanced and emerging market and developing economies. Policies that stimulate research and development, entrepreneurship, technology transfer, education, and public infrastructure investment can help lower the relative price of investment goods and boost investment.

### Key patterns over time
- Since the 1990s, capital goods prices relative to consumption prices have displayed two key patterns:
  - The four main types of fixed capital assets—structures, machinery and equipment (excluding transportation), transportation equipment, and intellectual property products—have evolved differently. The prices of machinery and equipment and transportation equipment have declined significantly since the early 1990s when compared with the consumption deflator.
  - Within tangible tradable capital goods, computing equipment and communications equipment experienced the largest price declines: computing equipment prices fell by 90 percent since 1990, and communications equipment prices fell by almost 60 percent.
- By 2014, compared with the early 1990s, the price of machinery and equipment declined relative to the consumption deflator in:
  - all advanced economies,
  - 87 percent of emerging market economies,
  - 68 percent of low-income developing countries.

### Cross-country patterns
- Despite broad-based declines over time, prices of tradable capital goods vary substantially across countries, especially relative to the price of consumption.
- ICP 2011 data indicate the absolute price of machinery and equipment was inversely related to countries’ development levels, with lower-income countries facing slightly higher prices than advanced economies; the same basket costs about 8 percent more in the median low-income country than in the median advanced economy.
- The higher relative prices in emerging market and developing economies reflect lower efficiency in producing investment and tradable goods and significantly higher trade costs, including tariffs.

*Source: Introduction, Chapter 3, "The Price of Capital Goods: A Driver of Investment Under Threat?" — World Economic Outlook, April 2019.*

### 2.7 times the price in the median advanced economy

### 2.7 times the price in the median advanced economy

### Key empirical findings on absolute and relative prices of machinery and equipment
- The absolute price of machinery and equipment is the price level of machinery and equipment relative to its US level, derived by the ICP using a similar basket of products across countries (ICP 2011).
- The relative price is the price of machinery and equipment relative to the price of consumption (ICP 2011).
- Regression relationships reported in Figure 3.3:
  - Absolute price: y = –0.025***x + 1.46
  - Relative price: y = –0.54***x + 7.30
- Relative to the price of consumption, the prices of machinery and equipment are significantly higher in emerging market and developing economies than in advanced economies.
- Lower-income countries also face marginally higher absolute prices of machinery and equipment.
- Comparable cross-country data on the price of capital goods are extremely scarce; ICP rounds (1985, 1996, 2011) are the key source cited for cross-country comparisons.

### Simple theoretical framework: why relative prices matter for investment
- A decline in the price of capital goods relative to other prices increases the attractiveness of investment versus consumption and thus leads to higher real investment rates (higher ratio of real investment to real output), holding other factors constant.
- Investment decisions are influenced by many other factors, including:
  - expectations of economic prospects;
  - availability and cost of finance;
  - the quantity of capital already in use relative to desired capital stock;
  - rate of depreciation of capital goods;
  - agents’ impatience.
- The relative price of capital goods is shaped by:
  - efficiency in producing machinery and equipment versus other sectors (relative productivity);
  - prices charged by international suppliers when countries import investment goods;
  - trade and transportation costs;
  - efficiency of domestic distribution sector;
  - import tariffs, customs regulations, and logistics time and cost;
  - tax policies (accelerated depreciation, investment tax credits, subsidies);
  - extent of corruption.

### Drivers of higher relative investment prices across countries
- Major capital goods exporters analyzed: United States, China, Germany, France, Japan.
- Unit-value analysis of capital goods imports from these exporters finds little systematic correlation between exporter unit values and per capita income of the importing country (Figure 3.4, panel 3).
- Trade costs exhibit a clear pattern of being much lower for advanced economies.
- Emerging markets and low-income developing countries face systematically higher policy-related and natural trade barriers:
  - policy-related barriers: higher tariffs, documentary and border compliance costs;
  - natural barriers: greater distance to exporters and weaker liner shipping connectivity.
- Trade-cost measures shown in Figure 3.5 (median and interquartile range by country group) include:
  - Distance to machinery and equipment exporters (Thousand kilometers)
  - Liner shipping connectivity (Index)
  - Freedom to trade internationally (Index)
  - Tariffs on capital goods (Percent)
  - Costs to import (Thousands of US dollars a container)
  - Time to import (Days)
- Empirical correlations (Figure 3.6):
  - The absolute price of machinery and equipment in 2011 was higher in countries with larger trade costs.
  - Panel 1 (Figure 3.6) reports percent change in the 2011 ICP absolute price of machinery and equipment associated with a one standard deviation increase in alternative measures of trade costs (basis points).
  - Panel 2 (Figure 3.6) decomposes cross-country variation in the 2011 ICP relative price into shares explained by:
    - differences in labor productivity in tradable goods sectors relative to nontradable sectors;
    - alternative measures of trade costs.
  - Trade costs and relative tradable productivity can together explain up to 60 percent of the cross-country variation in the relative price of machinery and equipment (depending on the trade-cost measure used).
- Policy-related trade barriers (tariffs, cost and time of importing) are more powerful predictors of relative prices than natural barriers (distance, connectivity).
- Evidence on exporter market shares and dynamics:
  - Emerging market and developing economies accounted for about one-third of global exports of capital goods in 2016, up from 5 percent in 1990.
  - China’s share rose from 1–2 percent in the 1990s to 18 percent in 2017.

### Over-time drivers of the large decline in relative prices (1995–2011 / 2000–11 analysis)
- Two-step empirical approach using sectoral producer price data across 40 advanced and emerging market economies (1995–2011):
  1. Estimate elasticity of producer prices to changes in sectoral labor productivity and exposure to international trade (import penetration = ratio of imports to domestic value added). Controls: country-year factors and time-invariant country-sector differences.
  2. Combine estimated elasticities with observed changes in relative labor productivity and trade exposure of the capital goods sector to quantify contributions to the decline in relative prices of machinery and equipment during 2000–11.
- Instrumental-variable strategy: import tariffs are used as an instrument for import penetration to isolate policy-driven changes in trade exposure.
- The decomposition separates contributions from:
  - trade-related changes in labor productivity (via exposure to competition/imports);
  - productivity changes due to other factors (sectoral technological advances).
- Findings indicate both productivity improvements in capital-goods-producing sectors and deepening trade integration contributed to the large declines in relative prices over the past 30 years.
- Caveats: endogeneity of trade exposure; tariffs as instrument do not fully address all endogeneity concerns.

### Implications and interpretation
- Cross-country and over-time evidence are consistent with the interpretation that higher relative prices of capital goods in emerging market and developing economies arise from:
  - higher trade barriers (policy and logistics);
  - lower productivity in production of capital goods and tradable goods more broadly.
- Given the role of relative prices in investment decisions, persistent higher relative prices of capital goods can impede real investment in lower-income countries.
- The analysis underscores policy levers that can reduce relative investment prices: lowering tariffs and other trade barriers, improving logistics and shipping connectivity, and raising productivity in tradable-goods sectors.

*Source: Chapter 3, “The Price of Capital Goods: A Driver of Investment Under Threat?” (International Monetary Fund, April 2019).*

### CHAPTER 3

### CHAPTER 3

### Trade, Productivity, and Sectoral Producer Prices
- Econometric analysis (see Online Annex 3.5 and Table 3.1) confirms:
  - A 1 percent increase in the import ratio (achievable by a 0.7 percentage point cut in tariffs) reduces the sectoral producer price by about 0.5 percent.
  - A 1 percent increase in sectoral labor productivity reduces producer prices by about 0.3 percent.
- Policy-induced changes in import penetration have a strong positive effect on labor productivity at the sector level; labor productivity of the capital-goods-producing sector is particularly sensitive to deepening trade integration.
- Illustrative policy magnitude:
  - If low-income countries were to bring capital goods tariffs to the level of those in advanced economies (reduce tariffs by roughly 8 percentage points), the price of investment goods would decline by about 16 percent.
    - Roughly 40 percent of that decline would come from the direct trade integration effect and the rest from higher productivity in the capital goods sector due to greater import competition.

- Key coefficients from Table 3.1 (sectoral regressions; all regressions include country-year and country-sector fixed effects; standard errors in parentheses):
  - Relative Import Penetration t – 1:
    - Column (1) OLS: –0.135*** (0.033)
    - Column (2) OLS: –0.107*** (0.037)
    - Column (3) IV: –0.574*** (0.163)
    - Column (4) IV (Relative Productivity): 1.363*** (0.363)
  - Difference for Capital Goods Sectors:
    - Column (1): –0.191** (0.081)
    - Column (2): 0.033 (0.322)
    - Column (4): 1.407** (0.671)
  - Relative Productivity t – 1 (columns reporting this): –0.316*** (0.035); –0.314*** (0.035); –0.328*** (0.032)
  - Relative Import Penetration for Capital Goods Sectors (additional reported coefficients):
    - –0.298*** (0.071); –0.541* (0.287); 2.770*** (0.564)
  - Number of Observations: 16,077 (for all columns)
  - Reported R^2: 0.62; 0.62; 0.56; 0.91

- Decomposition of decline in relative price of machinery and equipment (2000–2011) into four parts:
  1. Direct effect of deepening trade integration.
  2. Effect of trade integration through higher labor productivity.
  3. Effect of higher labor productivity not due to deepening trade integration.
  4. Residual (other factors).
- Findings from the decomposition:
  - Deepening trade integration accounts for the bulk of the decline in relative prices of machinery and equipment, both directly and indirectly via higher labor productivity in domestic capital goods producers.
  - Productivity gains in capital-goods-producing sectors not directly linked to trade integration are also significant.
  - A nontrivial portion of the decline in the price of investment goods—especially in emerging market and developing economies—can be attributed to other factors (for example, the downward trend in world interest rates, financial liberalization, and the emergence of China as a key supplier of tradable investment goods during the period).

### Macroeconomic Effects: Model Simulations
- Structural model used: IMF’s Global Integrated Monetary and Fiscal model (GIMF).
- Model experiments indicate:
  - Shocks to the relative productivity of the investment-goods-producing sector and tariff cuts that permanently lower the price of capital goods imports lead to sizable and long-lasting increases in the real investment rate in a representative emerging market economy.
  - Shocks that result in a 1 percent decline in the price of investment relative to consumption lead to a roughly 0.8 percent increase in the ratio of real investment to real GDP in the medium term.
  - For an average emerging market and developing economy with a ratio of real investment to real output of about 22 percent, a 1 percent decline in the relative price of investment would lead to an increase in the investment rate to 22.2 percent.

- Figure 3.8 comparison (simulated/estimated elasticities of real investment-to-GDP ratio to relative price of capital goods):
  - GIMF technology shock: (depicted elasticity)
  - GIMF capital goods tariff: (depicted elasticity)
  - Cross-country regression: (depicted elasticity)
  - Sectoral regression: (depicted elasticity)
  - All approaches deliver broadly consistent estimates of the elasticity.

### Empirical Evidence — Cross-Country and Sectoral Analyses
- Cross-country analysis:
  - Data: Over 60 years across 180 advanced and emerging market and developing economies from the Penn World Table database.
  - Estimation framework: Reduced-form relating real investment in machinery and equipment (share of real output) to price of machinery and equipment relative to price of consumption, using five-year averages and controlling for many global and country-specific factors.
  - Main empirical result (Table 3.2 and robustness checks):
    - A 1 percent decline in the relative prices of tradable capital goods is associated with a 0.3–0.5 percent increase in the real investment rate over a five-year period.
    - These estimates are robust to alternative specifications, focusing on post-1990, EMDEs-only, and alternative IV strategies.
    - Caveat: These empirical estimates likely represent an upper bound because relative investment prices are endogenous and reflect many factors, including policy changes that could directly impact investment rates.

  - Key coefficients from Table 3.2 (log real investment-to-GDP ratio regressions; IV with lagged price instrument; standard errors in parentheses):
    - Column (1) All: Log Relative Price coefficient = –0.377*** (0.116); Number of Observations = 658; Number of Countries = 127; R^2 = 0.41; First Stage F-Statistic = 118.80
    - Column (2) Post-1990: –0.292* (0.171); Observations = 553; Countries = 127; R^2 = 0.36; First Stage F-Statistic = 81.81
    - Column (3) EMDEs: –0.491*** (0.161); Observations = 457; Countries = 93; R^2 = 0.38; First Stage F-Statistic = 64.04

- Sectoral analysis:
  - Data: EU and World KLEMS covering 18 economies over 1971–2015; measures constructed for 15 broad economic sectors.
  - Baseline estimation: Machinery and equipment investment as share of sectoral real value added related to sector-specific relative prices, using five-year averages and country-period and country-sector fixed effects.
  - Estimated elasticity: A 1 percent decline in the relative price of machinery and equipment is associated with a 0.2–0.5 percent increase in the real investment rate in these capital goods—comparable to cross-country estimates.
  - Additional findings:
    - Declining investment prices are linked to higher output in the sector and marginally higher labor productivity.
    - Firm-level evidence from Colombia confirms that lower capital goods prices following a sizable tariff cut after trade reform in 2011 prompted firms to boost investment.

### Quantitative Contributions to Historical Investment Growth
- Illustrative decomposition (Figure 3.9) using cross-country elasticity and post-1990 relative price changes partitions the increase in real investment rates into:
  1. Decline in real investment prices.
  2. Change in relevant policies.
  3. Other factors (global trends in investment, convergence, growth expectations).
  4. Residual.
- Between 1990–94 and 2010–14:
  - Real investment-to-real GDP ratios in machinery and equipment grew by approximately 60 percent.
  - A significant portion of this increase can be explained by the precipitous fall in the relative price of machinery and equipment.
  - Improvements in policies and policy frameworks also contributed significantly to the rise in real investment in machinery and equipment in both AEs and EMDEs.
  - The decline in relative prices of tradable capital goods explains a sizable share of the increase in investment in tradable capital goods in both advanced and emerging market and developing economies.

*Source: CHAPTER 3, The Price of Capital Goods: A Driver of Investment Under Threat?, IMF, April 2019.*

### CHAPTER 3

### CHAPTER 3

### Summary and Policy Implications
- Declines in the relative price of tradable capital goods have been an important driver of stronger investment in emerging market and developing economies over the past three decades, alongside improved macroeconomic frameworks, synchronized global activity up to the global financial crisis of 2008–09, and falling global real interest rates.
- Empirical evidence and structural model simulations indicate the relative price of investment goods is an important driver of real investment rates in both advanced and emerging market and developing economies.
- A nontrivial share of the rise in real investment rates in machinery and equipment in both groups of economies can be attributed to the dramatic fall in the relative price of these goods over the past three decades.
- Sectoral analysis attributes the significant decline in the price of machinery and equipment mainly to:
  - faster productivity growth in the capital-goods-producing sector; and
  - deepening trade integration, which bolstered price competition and improved production efficiency in the investment goods sector.
- Policy implications:
  - Slowing trade liberalization since the mid-2000s, and potential reversal in some advanced economies, could weaken the tailwind to machinery and equipment investment provided by falling capital goods prices.
  - Policies to reduce trade costs and reinvigorate international trade would help maintain declines in relative capital goods prices and spur investment—especially regulatory cooperation, e-commerce facilitation, and leveraging complementarities between investment and trade.
  - Many emerging market and developing economies maintain tariff and other trade barriers that raise the relative price of investment: effective import tariffs on capital goods in 2011 were about 4 percent in emerging market and 8 percent in low-income developing countries, compared with close to zero in advanced economies.
  - Fully implementing commitments under the World Trade Organization’s Trade Facilitation Agreement could reduce non-tariff barriers by an equivalent of a 15-percentage point tariff cut in less-developed economies.
  - Continued technological advances and innovation in the capital-goods-producing sector—supported by policies encouraging research and development, entrepreneurship, technology transfer, education, and public infrastructure—can lower relative investment prices and generate dividends beyond aggregate productivity gains.
  - Policymakers should address distributional consequences and job disruptions from declining relative investment prices through long-term investment in education, lifelong skill upgrading, and policies facilitating reallocation of displaced workers to new jobs.

### Box 3.1 — The Price of Manufactured Low-Carbon Energy Technologies (key findings)
- Between 2009 and 2017:
  - prices of solar photovoltaics fell by 76 percent;
  - prices of onshore wind turbines fell by 34 percent.
- Global renewable energy capacity grew by about 6.5 percent a year between 2000 and 2017 and captured more than two-thirds of global investment in new generation capacity in recent years.
- Investment composition:
  - Hydropower dominated renewable investment up to 2008;
  - Wind technologies led in 2009;
  - Solar photovoltaics became the most popular investment choice in 2016;
  - In 2017, more was invested in solar photovoltaics than in all other low-carbon sources combined.
- Divergent cost paths:
  - Nuclear energy and hydropower costs rose by 21 percent and 9 percent, respectively, between 2009 and 2017.
  - Reasons include lumpiness of investment, limited component standardization, construction delays, and stricter local environmental and safety concerns for large-scale civil engineering projects.
  - Rapid cost declines in solar and wind reflect research and development, standardization, economies of scale, and learning by doing—factors that bode well for cheaper electric batteries through similar efficiencies.

### Box 3.2 — Evidence from Big Data: Capital Goods Prices across Countries (key findings)
- The Billion Prices Project database was used to compare online prices for 674 distinct Apple products across 27 economies, monthly from 2009 to 2012, normalized by US prices.
- Findings:
  - Absolute prices of identical Apple products differ significantly across countries and show no clear correlation with per capita income.
  - Relative to the overall GDP price level, the relative prices of capital goods tend to be significantly lower in richer countries, confirming regularities established with International Comparison Program data.
- Interpretation:
  - Online retail price differences reflect markups, local taxes and subsidies, transportation costs, and tariffs and other nontariff barriers.
  - The results support the view that richer economies face lower relative prices of capital goods.

### Box 3.3 — Structural Model Simulations: Sources of Investment Price Declines (key scenarios and quantitative impacts)
- Framework:
  - The IMF’s Global Integrated Monetary and Fiscal model is used to study medium-term macroeconomic effects in a small emerging market economy of two scenarios that produce an identical long-term 1 percent decline in the relative domestic price of investment goods (relative to the consumer price index).
- Scenarios:
  1. Investment-specific technological change (higher domestic efficiency in producing new capital): interpretable as greater international diffusion of technological know-how, lower domestic costs in capital goods sectors, improved organizational efficiency, or similar productivity improvements.
  2. Permanent decline in tariffs on imported capital goods: a 4 percentage point permanent decline in investment goods tariffs is required to obtain the 1 percent long-term decline in the relative price of capital given the model’s assumed share of capital goods imports in overall investment spending (about 33 percent). This tariff cut entails a recurrent fiscal cost of about 0.25 percent of annual GDP.
- Medium-term (10 years after the shock) simulation results:
  - Both scenarios increase returns to capital by similar amounts and trigger similar increases in investment.
  - Effects on real output differ:
    - Technology shock: real output increases by 0.7 percent of GDP.
    - Capital goods tariff reduction: real output increases by 0.5 percent of GDP.
  - Difference arises because permanent improvement in the efficiency of newly produced capital expands the production possibility frontier, producing permanent increases in household income and consumption.
- Additional notes:
  - A 1 percent decrease in investment goods prices caused by a decrease in markups in the investment goods sector produces very similar effects to an increase in investment-specific productivity.
  - The required decrease in tariffs reflects real exchange rate depreciation observed in the tariff scenario.

*Source: CHAPTER 3 — THE PRICE OF CAPITAL GOODS: A DRIVER OF INVESTMENT UNDER THREAT?, International Monetary Fund | April 2019*

### Box 3.3. On the Underlying Source of Changes in Capital Goods Prices: A Model-Based Analysis

### Box 3.3. On the Underlying Source of Changes in Capital Goods Prices: A Model-Based Analysis

### Model-based comparison of tariff-reduction scenarios
- Mechanism when tariffs on capital goods fall:
  - Lower capital goods prices act like a subsidy, making investment cheaper.
  - Tariff revenue forgone creates a government revenue shortfall, resolved by lowering public transfers to households.
  - Lower public transfers generate a headwind to private consumption.
  - Higher relative demand for imports places pressure on the real effective exchange rate to depreciate, further raising the domestic consumer basket price and weighing on consumption.
  - Net effect: little increase in consumption despite cheaper investment.
- Supply-side effects in both scenarios stem largely from lower investment costs.
- Simulation comparing an investment-specific tariff cut with an equivalent-fiscal-revenue general tariff cut:
  - Investment-specific tariff scenario:
    - Investment increase: 1.34 percent
    - Output increase: 0.5 percent
  - General tariff decrease (equivalent in fiscal revenue terms):
    - Investment increase: 0.23 percent
    - Output increase: 0.18 percent
  - Interpretation: when the domestic relative price of investment does not decline, the investment and output responses are much smaller.
- Policy implication from model results:
  - Lowering barriers that hamper trade in capital goods and promoting research and development that improve efficiency in the capital goods sectors are good for output, investment, and consumption in the long term, even if they entail some fiscal costs.

### Firm-level evidence from Colombia (tariff reform 2011)
- Context and reform:
  - Colombian tariff reform aimed to simplify the tariff structure and boost economic growth.
  - Between 2010 and 2011, the average tariff rate on imported goods declined by close to 4 percentage points, from 12.5 percent in 2010 to 8.7 percent in 2011.
- Empirical approach:
  - Event study relating change in firm-level investment rate (investment over total fixed assets for a given firm i) before and after the tariff reform to changes in:
    - ΔCapital Input Tariff_s(i)
    - ΔOther Input Tariff_s(i)
    - ΔOutput Tariff_s(i)
  - Output Tariff_s(i) is the simple average of most-favored-nation tariffs across Harmonized System six-digit products within the 33 sectors, s(i), intended to capture competition effects.
  - Capital Input Tariff_s(i) and Other Input Tariff_s(i) follow Amiti and Konings (2007): weighted averages of output tariffs in all capital goods and other sectors, using 2007 input-output table shares as weights to reflect input usage.
  - Data sources: investment from Encuesta Anual Manufacturera; tariff data from Felbermayr, Teti, and Yalcin (2018).
  - Use of fixed sector-level input-output matrices reduces endogeneity concerns from firm-level input-output matrices.
- Estimated effects on investment (Figure 3.4.2):
  - A 1 percentage point reduction in capital goods input tariffs is associated with a 0.4 percentage point increase in investment (statistically significant at the 1 percent level).
  - A 1 percentage point reduction in noncapital input tariffs is associated with a 0.15 percentage point increase in investment (statistically significant).
  - Reductions in output tariffs are not associated with significant changes in firms’ investment decisions in the short term.
- Interpretation and robustness:
  - Results echo model simulations showing a larger investment response to capital goods tariff cuts than to general tariff cuts.
  - Coefficients on input tariff changes can be interpreted as price effects on investment under the assumption that tariffs are fully passed into importers’ prices; partial pass-through would attenuate estimates toward zero relative to the true price effect.
  - Results are robust to inclusion of standard controls (such as firm size or sales growth) and to using wider time windows around the tariff cuts.
- Empirical conclusion:
  - Firm-level evidence from a recent trade reform in a large emerging market economy shows firms’ investment choices are sensitive to the price of capital goods.

### Authors and methodological notes
- The authors of the box are Sergii Meleshchuk and Yannick Timmer.
- Notes on statistical significance: reported point estimates include significance annotations (***p < 0.01; **p < 0.05; *p < 0.1) as shown in Figure 3.4.2.

*Source: Box 3.3, World Economic Outlook: Growth Slowdown, Precarious Recovery (April 2019).*

---


_Source: https://www.imf.org/-/media/files/publications/weo/2019/april/english/ch3.pdf_