## _sdn1503

## Source details

**Canonical URL:** [_sdn1503](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2015/_sdn1503.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2015/_sdn1503.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2015/_sdn1503.pdf.json)

---

### EXECUTIVE SUMMARY — Main findings
- Trend growth in advanced economies has been slowing, driven by a slowdown in human and capital accumulation and declining total factor productivity (TFP).
- The decline in TFP before the crisis reflected both reallocation of resources to sectors with slower productivity growth and declining productivity growth within those sectors.
- Policy distortions are a major source of inefficient resource allocation across sectors; agriculture and personal services often show the most distortions.
- No one-size-fits-all policy prescription; reform priorities depend on country-specific settings and distance from the technology frontier.
- Product market reforms, especially in services, can yield significant and increasing productivity gains over time and support job creation.
- Investments in knowledge capital and innovation (labor skills development, ICT capital, and R&D) boost productivity, particularly in countries closer to the frontier; infrastructure yields significant productivity and more immediate employment/output benefits.
- Labor market reforms have mixed and complex productivity effects; reducing labor tax wedges is associated with higher productivity growth over time, while labor reforms tend to matter more for employment than for productivity.
- Reforms can have short-term employment and output costs; compensation mechanisms and timing matter, but some reforms show positive short-run effects even in weak demand environments.

### I. CONTEXT
- Analysis period: sector-level productivity developments in advanced economies over 1970 to 2007.
- Motivation: protracted recoveries since the 2008–12 balance sheet recessions despite large policy stimulus, raising questions about potential output and sources of growth.
- Conceptual point: as countries approach the technology frontier, catch-up gains from capital deepening, learning, and structural transformation diminish.
- Data source: harmonized industry-level dataset across advanced economies (EU KLEMS underlying), with caveats on measurement and coverage.

### II. GROWTH AND STRUCTURAL TRANSFORMATION — Aggregate and drivers
- Aggregate trends:
  - Output growth declined from an annual average of more than 3½ percent in the 1970s to less than 2½ percent in the mid-2000s.
  - Trend growth in output per hour worked declined from about 3 percent annually in 1970 to less than 1½ percent in 2007.
  - Trend estimates derived using an HP filter with a smoothing parameter of 100 on the growth series.
- Cross-country heterogeneity:
  - Slowdown most marked in core Europe and euro area countries with high borrowing spreads.
  - United States experienced an ICT-driven productivity surge primarily from 1995–2003 that was waning by the crisis.
  - Japan had catch-up growth through the 1980s but trend slowed by the late 1990s; United Kingdom exhibited faster labor productivity growth from the mid-1990s.
- Labor supply and demographics:
  - Labor supply issues did not significantly slow long-run growth relative to productivity contributions; rising dependency ratios were a drag, especially in Japan.
  - Labor utilization contributed less to the drag after 1990, reflecting rising female participation in some countries.
- Capital deepening and human capital:
  - Decline in growth of human and physical capital per worker since the 1990s contributed importantly to labor productivity decline.
  - ICT-capital deepening retreated from the mid-1990s peak; non-ICT capital deepening also declined, reflecting in part decreases in public investment.
  - Educational attainment appears close to plateauing in most economies.
- TFP:
  - TFP growth stagnated or declined in most developed economies even before the crisis and fell sharply during the crisis.
  - TFP growth fell most dramatically in euro area countries with high borrowing spreads, averaging a half percentage point decline every decade.
  - United States’ TFP slowdown started well before the crisis; Japan saw a TFP rebound from negative territory in the 1990s.

### Technological progress versus efficiency (SFA) — Summary findings
- Method:
  - Stochastic frontier analysis (SFA) estimated a TFP frontier across advanced economies and distance-from-frontier (technical efficiency) for each country.
  - Frontier represents maximum possible TFP under current technologies; distance-from-frontier represents domestic efficiency in implementing global technological know-how.
- Key findings:
  - “Advances in both the technology frontier and the pace of catch-up to the frontier stalled.”
  - Rolling window SFA for 1983–2007 indicates the growth rate of the productivity frontier declined “at about a third of a percentage point per decade.”
  - Estimated technical efficiency stalled, with the average country moving away from the frontier.
- Sectoral dynamics:
  - Services now account for “more than three-quarters of the total labor share.”
  - Within services, productivity growth was “much lower than in the rest of the economy” for nonmarket and personal services.
  - Distribution and ICT services experienced falling prices and rising labor productivity, gaining real value-added share while shedding labor.
  - ICT goods and services led sectoral TFP growth in 2000–07, especially in the United States and Nordic countries.
  - Aggregate TFP decline prior to the crisis driven by reallocation toward lower-growth sectors and declining within-sector productivity in sectors now accounting for most employment.
  - Sectoral TFP level gaps are large; the frontier economy varies by sector (e.g., Germany led in manufacturing; the United States led in finance, business, and personal services).
- Policy scope:
  - Large sectoral TFP gaps point to room for policy and structural reforms to improve allocative and productive efficiency; frontier sectors require innovation in products and processes.

### Measuring misallocation across sectors — Methodology and quantitative results
- Methodology:
  - Framework builds on Aoki (2012) using EU KLEMS sector-level value-added, inputs, and payments for 12 advanced economies over 2000–07.
  - Static general equilibrium model with frictions represented as wedges affecting effective cost of capital and labor; counterfactuals compute Pareto efficient allocation holding technology, aggregate labor and capital, and preferences constant.
  - Labor measured as total hours worked by employees (baseline) and alternatively as a skill-weighted labor index.
- Sectors with highest distortions:
  - Agriculture often “favored” (positive labor and capital wedges), notably in Japan and Portugal — likely due to subsidies or preferential policies.
  - Personal services usually exhibit higher labor distortions (low marginal returns to labor), possibly reflecting lack of competition or regulatory constraints.
- Cross-country heterogeneity:
  - Italy: positive capital wedge in business sector indicates over-accumulation of capital and low capital returns.
  - Portugal: low labor wedges in financial and ICTGS indicate barriers to entry and frictions limiting employment.
  - United States: misallocation driven by capital distortions, particularly in mining.
- TFP gains from eliminating distortions — Full adjustment (Baseline Pareto) TFP level gains (Percent):
  - Austria: 9.9
  - Australia: 7.2
  - Denmark: 9.8
  - Finland: 10.9
  - Germany: 3.8
  - Italy: 19.5
  - Netherlands: 7.4
  - Portugal: 13.7
  - Sweden: 4.4
  - Spain: 6.2
  - United Kingdom: 5.3
  - Japan: 9.2
  - United States: 6.7
- With labor skill composition (averages for 2000–07 or latest available) (Percent):
  - Austria: 6.3
  - Australia: 3.7
  - Denmark: 5.1
  - Finland: 5.2
  - Germany: 1.9
  - Italy: 4.5
  - Netherlands: 5.3
  - Portugal: 12.0
  - Sweden: 2.1
  - Spain: 3.2
  - United Kingdom: 3.1
  - Japan: 4.5
  - United States: 3.2
- Decomposed contributions (Adjust Labor Allocation Only — Percent):
  - Austria: 3.3
  - Australia: 3.3
  - Denmark: 4.4
  - Finland: 5.4
  - Germany: 1.9
  - Italy: 14.3
  - Netherlands: 2.0
  - Portugal: 1.5
  - Sweden: 2.2
  - Spain: 2.9
  - United Kingdom: 2.1
  - Japan: 4.4
  - United States: 3.4
- Decomposed contributions (Adjust Capital Allocation Only — Percent):
  - Austria: 10.2
  - Australia: 8.7
  - Denmark: 10.7
  - Finland: 17.9
  - Germany: 8.1
  - Italy: 36.4
  - Netherlands: 12.1
  - Spain: 11.0
  - United Kingdom: 12.9
  - Japan: 13.7
  - United States: 6.2
  - (Table shows “-” for Portugal and Sweden in capital-only column)
- Key interpretation:
  - Average headline TFP gains across countries about 9 percent in the baseline full-adjustment scenario; range from 3.8 percent (Germany) to 19.5 percent (Italy).
  - Accounting for labor skill composition affects estimated gains; labor vs. capital reallocation contributions differ by country.
- Growth implications (Pareto efficient allocation achieved within 10 years):
  - Italy: potential increase in annual growth rates of TFP (and GDP) by 1.8 percent per annum over the decade.
  - Portugal: potential increase by 1.3 percent per annum over the decade.
- Caveat: exercise assumes complete elimination of underlying distortions and considers only cross-sector reallocation (a lower bound on overall gains).

### Drivers of TFP growth — Regulation, skills, ICT, and R&D
- Conceptual framework distinguishes imitation (adoption) from innovation; countries farther from frontier grow mainly via adoption.
- Structural variables considered:
  - Product market regulation stringency (industry-specific OECD Indicators).
  - Employment protection rigidity (OECD Employment Protection Legislation indicator).
  - Inputs: high-skilled labor intensity, ICT capital inputs, R&D spending.
- Empirical findings:
  - Services among the most heavily regulated industries, especially Italy and Japan; wide dispersion across peers.
  - United States leads peers in innovation indicators and high-skilled labor use.
  - In services, innovation often arises from tangibles (automation, ICT) and nontangibles (business processes) rather than R&D spending.
  - Productivity growth driven by catching-up (negative coefficient on TFP gap) and frontier developments (positive coefficient on frontier TFP growth), consistent with spillovers.
- Product market reforms:
  - Higher product market regulation in services associated with lower TFP growth; negative effect larger the closer the industry is to the frontier.
  - Example quantification: if Italy reduced product market regulation to Netherlands level (frontier in distribution services), TFP growth gain within services could be about 0.4 percent per year.
- Labor market rigidities:
  - Economy-wide employment protection indicator not significant in panel for industry-level TFP growth, reflecting institutional diversity and lagged reform effects.
  - Firm-level evidence: less stringent labor institutions and efficient human capital use facilitate labor movement to productive firms and creative destruction.
  - Mandatory dismissal regulations and long unemployment benefits can depress productivity where restrictions bind.
  - Policy implication: improving labor market institutions and reducing skill mismatches could enhance productivity via labor reallocation.

### Productivity impact of knowledge capital and innovation — Core findings
- Core findings:
  - More intensive use of high-skilled labor, ICT capital inputs, and higher R&D spending associated with higher productivity growth.
  - Channels:
    - High-skilled labor fosters skill-intensive industries and technology diffusion (example: increasing UK high-skilled share to Japan’s level could boost productivity growth by 0.7 percent per year).
    - ICT diffusion supports organizational and process efficiency.
    - R&D creates new efficiency-enhancing technologies.
- Varying sectoral effects and spillovers:
  - Human capital and R&D spur TFP growth in ICT-related industries (and services for skilled labor), strongest for countries closer to the frontier.
  - Trade and technology spillovers mean reforms in leaders generate positive spillovers elsewhere.
  - ICT use impact within services highest where sector productivity lags, highlighting role for ICT diffusion and removal of trade/FDI barriers.
- Quantified potential gains (closing gap between 25th percentile and best-practice level, median country):
  - Improvements in structural variables could generate additional TFP growth gains of ¼ to ½ a percentage point annually across all industries.
- Table 4 (TFP Growth Gains from Improvement in Structural Determinants — Percentage points, annually) — selected entries:
  - Product Market Regulations: Manufacturing -0.1; Services -0.1
  - Labor Skill: Manufacturing -1.4; Services 1.2; ICT 0.4
  - ICT Capital: Manufacturing 0.7; Services 0.7; ICT -0.3
  - R&D Expenditure: Services 1.9; ICT 0.2
  - Interpretation: product market regulation gains accrue largely to services; R&D yields largest gains in ICT; human capital and ICT use have broader effects.
- Identifying reform shocks:
  - Large reform definitions vary by indicator (e.g., R&D/ICT/high-skilled: change above two standard deviations; product/labor/infrastructure: change beyond one standard deviation).
  - Infrastructure shocks mainly in 1970s and since waned; product market reforms spurted in 1990s into 2000s; labor market reforms less frequent.
- Dynamic effects:
  - Productivity impact typically grows over time; short- and medium-term effects generally positive.
  - Examples:
    - A 10 percent shock in R&D spending: TFP up 0.7 percent after three years and 2 percent over five years.
    - An ICT capital shock: cumulative TFP level increase in ICT-related industries of about 30 percent after five years (two standard deviations of average cumulative five-year TFP change).
  - Labor-market nuance:
    - Reduction in employment protection associated with a decline in TFP after three years (possible weaker firm-specific skill investment).
    - Reduction in the labor tax wedge associated with higher productivity three years after reform.
- Heterogeneity:
  - Product market liberalization yields medium-term gains highest in ICT and personal services.
  - Lower tax wedges, higher R&D, and better human capital boost productivity in innovation-intensive manufacturing and ICT.
  - Infrastructure has positive productivity impact across countries, irrespective of frontier distance.
- Reform impact across the business cycle:
  - Infrastructure shocks yield higher productivity gains during downturns.
  - Product market liberalization and innovation shocks yield higher productivity gains during upturns.
- Dynamic effects on employment and output:
  - Short-term trade-offs: reforms can cause temporary negative effects on output and employment.
  - Employment outcomes (cumulative three-year and five-year):
    - Product market liberalization and infrastructure shocks: positive and persistent employment impact.
    - Intensive use of skilled labor: positive employment in manufacturing and ICT; lower employment in agriculture, utilities, mining.
    - Labor market liberalization (reduced dismissal costs): boosts employment in personal services short term.
    - ICT shocks: associated with lower employment in manufacturing and personal services (automation/displacement).
    - R&D shocks: associated with lower employment in distribution services.

### Impact on output — Magnitudes and timing
- Output gains materialize predominantly in the medium term:
  - Product market liberalization and more intensive use of high-skilled labor raise output across all industries by about 2 percent over the medium term.
  - Infrastructure shocks increase output across all industries by around 3 percent over the medium term.
  - Innovation shocks (R&D/knowledge capital) increase output across all industries by around 5 percent over the medium term.
- Labor market liberalization (lower employment protection) has a negative aggregate short-term impact on output (though not on employment).
- Infrastructure shocks have positive short- and medium-term output impacts across almost all sectors.
- Heterogeneity and interpretation:
  - Effects depend on country-specific policy and institutional settings; associations do not imply uniform causation.
  - Complementarities and sequencing of reforms across areas matter for realizing gains.
- Final policy implications (back to basics):
  - To increase growth potential, advanced economies must:
    - boost workforce productive capacity;
    - invest in physical and IT infrastructure;
    - encourage innovation;
    - remove binding product and labor market rigidities that impede allocation to dynamic sectors/firms;
    - augment human capital, reduce skills gaps, and improve education quality.
  - Government role: correct market failures in innovation, provide incentives for R&D, and coordinate large infrastructure projects.
  - Tailoring to conditions: where demand is weak and labor slack exists, prioritize product market reforms and investment in productive infrastructure to reap immediate employment and output benefits.

### Key takeaways and policy recommendations
- Services reform priority:
  - Services are the most important and most highly regulated sector across advanced economies; significant upside exists from product market reforms to increase competition in services.
  - Investment in information technology can substantially boost service productivity in lagging regions.
  - Labor market policies should avoid excessive regulation and extreme neglect of labor conditions.
- Spillovers:
  - Performance gains by technology leaders produce global spillovers supporting productivity elsewhere.
  - Increasing returns to scale in knowledge capital create opportunities for domestic and global gains, especially for countries far from the frontier.
- Infrastructure investment:
  - Transport, energy, and telecommunications investment improve productivity, boost employment, and increase output.
  - Debt-financed projects could have large productivity and output effects if needs are met through efficient investment.
- Improve resource allocation:
  - Reduce policy distortions (subsidies, exemptions, size-dependent policies) that skew resources toward low-value activities.
  - Rationalize agricultural subsidies and increase competition in personal services to reduce misallocation.
- Calibrate reforms:
  - Reform design must be granular and country-specific; no one-size-fits-all solution.
  - Sequencing and complementarities across reforms are critical to realize potential gains and mitigate short-run costs.

### Annex: methodological note (Stochastic Frontier Analysis)
- SFA concept: a world frontier technology exists for each industry/sector; country TFP depends on the world frontier TFP level and the efficiency of technology use.
- Model specification close to Pitt and Lee (1981); world frontier productivity level determined by a constant and a trend component.
- TFP level estimates link EU KLEMS TFP growth with Groningen Growth and Development Centre Productivity-Level database (PPP-adjusted industry TFP levels in 1997).
- Productivity series constructed for 23 sectors spanning 1970–2007.
- Estimation uses maximum likelihood on 10-year rolling window sectoral TFP data; trend of frontier TFP growth and average TFP level as percentage of world frontier reported for each sector.

*Source: IMF staff analysis in chapter excerpts from the provided IMF PDF content.*

### EXECUTIVE SUMMARY ____________________________________________________________________________ 4

### EXECUTIVE SUMMARY

### Main Findings
- Trend growth in advanced economies has been slowing, driven by a slowdown in human and capital accumulation and declining total factor productivity (TFP). The decline in TFP before the crisis reflected not only the reallocation of resources to sectors where productivity growth was slower, but also declining productivity growth within those sectors which increasingly account for the bulk of employment and economic activity.
- Policy distortions are a major source of inefficient resource allocation across sectors. In most advanced economies, the agriculture and personal services sectors have the most distortions, possibly reflecting the effect of subsidies, preferential policies, and the lack of competition.
- There is no one-size-fits-all policy prescription for all country circumstances and growth experiences. Reform priorities depend on country-specific settings, including the scale of particular policy distortions and the distance from the technology frontier.
- Significant payoffs can be realized with product market reforms, especially in some parts of the services sector, which is the largest sector overall. Moreover, productivity gains from these reforms increase over time. Liberalizing entry into regulated services sectors can also be an important source of job creation and output growth.
- Investments in knowledge capital and innovation (labor skills development, information communication technology capital, and research and development) can boost productivity, especially in countries closer to the technology frontier that generate large global productivity spillovers. The productivity payoffs from investments in infrastructure are significant, but there are more immediate employment and output benefits.
- The productivity impact of labor market reforms is more difficult to discern, reflecting the particular complexity of gauging policies in this area. There is some evidence that reducing labor tax wedges is associated with higher productivity growth over time. In general, labor market reforms tend to matter more for employment, with modest effects on productivity, suggesting the need to avoid either excessive regulation or excessive neglect of labor conditions.
- Reforms can have short-term costs in terms of employment and output, suggesting the need for compensation mechanisms. But there is some evidence of positive effects of reforms even in the short run, which is useful to consider when judging policy priorities at a time of weak demand.

### I. CONTEXT
- The study examines sector-level productivity developments in advanced economies over 1970 to 2007 to highlight trends already in progress before the global financial crisis.
- Motivation: recoveries since the 2008–12 balance sheet recessions have been protracted despite large policy stimulus, raising questions about potential output and future sources of growth.
- Key conceptual point: some deceleration in economic growth in advanced economies is expected as countries get closer to the technology frontier; catch-up gains from capital deepening, learning and technology transfer, and structural transformation diminish.
- Data source: the analysis is based on a unique harmonized industry-level dataset across advanced economies (EU KLEMS database noted as the underlying source), with caveats on measurement and country coverage at detailed industry level.

### II. GROWTH AND STRUCTURAL TRANSFORMATION
- Aggregate trends
  - Secular decline in output and productivity growth evident over the last four decades, visible in pre-crisis data and amplified by the global financial crisis.
  - Output growth declined from an annual average of more than 3½ percent in the 1970s to less than 2½ percent in the mid-2000s just before the crisis.
  - Trend growth in output per hour worked declined from about 3 percent annually in 1970 to less than 1½ percent in 2007.
  - Trend estimates noted as derived using an HP filter with a smoothing parameter of 100 on the growth series.
- Cross-country heterogeneity
  - Slowing labor productivity growth varies by country group: most marked in core Europe and in euro area countries with high borrowing spreads; the United States experienced a ICT-driven productivity surge primarily from 1995–2003 that was waning by the crisis; Japan saw catch-up growth through the 1980s but trend slowed by the late 1990s; the United Kingdom exhibited faster labor productivity growth from the mid-1990s.
- Role of labor supply and demographics
  - Labor supply issues (fertility, aging, labor utilization) did not significantly slow the long-run growth trend relative to productivity contributions.
  - The rising dependency ratio has been a drag for most countries, especially Japan which experienced a major increase in old-age dependency since the 1990s and negative population growth.
  - Labor utilization (participation) contributed less to the drag after 1990 than before, reflecting rising female participation in some countries.
- Capital deepening and human capital
  - Decline in growth of human and physical capital per worker since the 1990s contributed importantly to labor productivity decline.
  - ICT-capital deepening retreated from the mid-1990s peak, notably in the United States.
  - Non-ICT capital deepening also declined pre-crisis, reflecting in part a long-run decrease in public investment.
  - Educational attainment (average years of schooling) appears close to plateauing in most economies; for cohorts born after 1950 in the United States, rise in educational attainment slowed markedly and ceased for the most recent cohort (Fernald and Jones 2014).
- Total Factor Productivity (TFP)
  - TFP growth stagnated or declined in most developed economies even before the crisis; TFP growth fell dramatically during the crisis but was stagnant at the pre-crisis peak across economies.
  - TFP growth fell most dramatically in euro area countries with high borrowing spreads, averaging a half percentage point decline every decade.
  - The United States’ slowdown in TFP growth started well before the crisis; Japan experienced a TFP rebound from negative territory in the 1990s due to firm-level restructuring.
  - Natural resource economies experienced a big drop in aggregate TFP growth in the pre-crisis 2000s despite commodity booms.

*THE NEW NORMAL — EXECUTIVE SUMMARY, INTERNATIONAL MONETARY FUND*

### 12. Technological progress versus

### _sdn1503 - 12. Technological progress versus

### Technological progress versus efficiency (stochastic frontier analysis)
- Method:
  - Applied stochastic frontier analysis (SFA) to estimate a TFP frontier across all advanced economies and a distance-from-frontier (technical efficiency) for each country (see Annex 1).
  - Frontier represents maximum possible TFP levels under current technologies; distance-from-frontier represents domestic efficiency in implementing global technological know-how.
- Key findings:
  - “Advances in both the technology frontier and the pace of catch-up to the frontier stalled.”
  - Rolling window estimates of the SFA model for 1983–2007 indicate the growth rate of the productivity frontier declined “at about a third of a percentage point per decade” (the blue line in Figure 5).
  - Estimated technical efficiency stalled over time, with the average country moving away from the frontier (the red line in Figure 5).
  - These trends warrant further examination of factors underlying TFP growth.

### Sectoral productivity dynamics
- Structural shifts and implications:
  - Over the last four decades advanced economies shifted away from agriculture and manufacturing into services; services now account for “more than three-quarters of the total labor share.”
  - Structural transformation can raise growth if driven by technical change and efficient factor allocation; however, globalization, changing consumer preferences, and ICT-driven structural shifts have different effects across countries.
- Shifts within services (1980–2007):
  - Common patterns across advanced economies:
    - Increases in value-added and employment shares in finance and business services (pre-crisis boom), especially pronounced in the United Kingdom.
    - Gains in personal services (hotels, restaurants, social and personal services) and nonmarket services (public administration, education, health, and real estate).
    - Productivity growth in these services was “much lower than in the rest of the economy,” consistent with Baumol-style low innovation scope.
    - Distribution and ICT services experienced falling prices and rising labor productivity growth, gaining real value-added share while shedding labor.
- TFP within sectors:
  - Aggregate TFP decline prior to the crisis was driven both by resource reallocation to lower-productivity-growth sectors and, importantly, by declining productivity growth within sectors that now account for most employment and activity.
  - Figure 7 compares 1990–99 and 2000–07 average TFP growth across sectors and countries; relatively protected services (nonmarket, personal, business services) were heavy drags on TFP growth in many countries and “strongly negative in Italy and Spain.”
- ICT dynamics:
  - ICT goods and services (ICTGS) led sectoral TFP growth in 2000–07, especially in the United States and Nordic countries; Japan, core Europe, and euro area countries with high borrowing spreads lagged.
  - Slower ICT adoption and diffusion in services in Europe (Finland and Sweden exceptions) is cited as a key driver of the U.S.–Europe productivity gap.
  - Evidence that ICT-driven productivity gains may have peaked: slowdown of TFP growth in IT-producing sectors in the United States in advance of the crisis (Fernald 2014); commentators suggest frontier may return to slower growth rates last seen in the early 1990s (Gordon 2012).
- Distribution services:
  - Distribution services benefited from ICT innovations and recorded above-average TFP growth in most economies.
  - The United States and some euro area countries with high borrowing spreads saw marked TFP slowdowns in distribution on the eve of the crisis.
- Other boom and lagging sectors:
  - Finance registered high measured productivity pre-crisis in the United Kingdom and some European countries; durability of these gains is uncertain given crisis events.
  - Construction and real estate showed disappointing TFP growth prior to the crisis, notably in the United States, Japan, and euro area countries with high borrowing spreads.
  - Extractive industries’ productivity growth slowed prior to the crisis, possibly due to pursuing more marginal extraction amid booming commodity prices.
- Sectoral TFP level gaps:
  - Large cross-country variation in sectoral TFP levels prior to the crisis; even technologically advanced countries can have sectors with TFP levels “more than half that of the leader.”
  - The frontier economy varies by sector (e.g., Germany led in manufacturing but lagged in services; the United States led in finance, business, and personal services but lagged in non-ICT manufacturing).
  - Caveats: natural advantages, agglomeration, and measurement issues (industry misclassification, cross-country comparability of hours worked and capital services) affect level comparisons.
- Policy scope:
  - Large gaps in sectoral TFP levels point to room for policy and structural reforms to facilitate catch-up by improving allocative efficiency and productive efficiency.
  - For countries at or near the production frontier in certain sectors (examples cited include the United States in personal services, the Netherlands in distribution), further gains may require innovations in products and processes.

### Resource allocation and productivity
- Aggregate dependence on allocation:
  - Aggregate TFP depends on both within-firm/industry efficiency and how inputs are allocated across firms and industries; reallocation of capital and labor toward more productive firms is a key growth mechanism.
- Misallocation and impediments:
  - Misallocation arises when impediments block factor movement across heterogeneous firms and sectors, producing persistent rate-of-return differentials and undermining aggregate TFP.
  - Examples of distortions:
    - Exemptions from regulations or taxes, heavily subsidized agriculture, “size-dependent policies” (e.g., limits on retail establishment size in Japan; employment protection policies in Italy and France that take effect beyond certain size thresholds).
    - Labor and product market rigidities, market structure, imperfect competition.
    - Financial frictions increasing dispersion in borrowing costs and capital misallocation (Gilchrist, Sim, and Zakrajsek 2013).
  - Such distortions lead to inefficient capital and labor choices, lowering aggregate TFP (Restuccia and Rogerson 2008; Hsieh and Klenow 2009).
- Within-sector misallocation gains:
  - Substantial TFP gains can be realized by reducing misallocation within narrowly-defined industries.
  - Evidence:
    - Dias, Marques, and Richmond (2014): better allocation across firms within services industries in Portugal could boost measured TFP by “about 40 percent.”
    - Bartelsman, Haltiwanger, and Scarpetta (2013): idiosyncratic distortions suggest output could be improved “by as much as 15 percent in some European countries” by improving resource allocation across manufacturing establishments.
  - Historical evidence: Baily, Hulten, and Campbell (1992) attribute 50 percent of manufacturing productivity growth in the U.S. in the 1970s–1980s to factor reallocation and entry/exit; Barnett and others (2014) find labor reallocation explained 48 percent of labor productivity growth in many U.K. sectors in the five years prior to 2007.
- Policy implications (implied by analysis):
  - Reduce distortions that impede factor reallocation (tax/regulation exemptions, subsidies, size-dependent policies).
  - Address labor and product market rigidities and market structure issues to facilitate firm entry, exit, and scaling.
  - Mitigate financial frictions to reduce dispersion in borrowing costs and encourage efficient investment in technology-enhancing capital.
  - Promote reforms that improve allocative and productive efficiency within sectors to capture substantial TFP gains.

*Source: IMF staff analysis in “Technological progress versus efficiency” (chapter 12) from the provided IMF PDF content.*

### 26. Measuring misallocation across sectors. While within-industry misallocation accounts for

### 26. Measuring misallocation across sectors

### Methodology and counterfactual exercise
- Framework:
  - Builds on Aoki (2012) and uses sector-level value-added outputs, factor inputs, and input payments from the EU KLEMS database for 12 advanced economies over 2000–07.
  - Static general equilibrium model with multiple sectors; frictions abstracted as wedges that affect effective cost of capital and labor (see Technical Appendix).
- Steps of the quantitative exercise (Box 1):
  - Measure sectoral TFP for each year from the data.
  - Estimate labor and capital frictions in each sector by measuring cross-sector differences in marginal returns to inputs; calibrate wedges so the model replicates observed allocations and outputs.
  - Compute counterfactual aggregate TFP with identical frictions across sectors (Pareto efficient allocation) holding technology, aggregate labor and capital stock, consumer preferences constant.
  - Also consider counterfactuals that change only capital frictions or only labor frictions to identify individual contributions.
- Labor input measurement:
  - Baseline: total hours worked by employees (no skill differentiation).
  - Alternative: labor index as weighted sum of hours by skill groups (accounts for labor skill heterogeneity).
  - Results reported for both baseline and skill-weighted scenarios.

### Sectors with the highest distortions
- General patterns:
  - Agriculture sector is often “favored” in most advanced countries: lower marginal returns to capital and labor (positive labor and capital wedges), notably in Japan and Portugal.
  - Underlying distortions in agriculture likely stem from subsidies or preferential policies.
  - Personal services sector usually exhibits higher labor distortions (low marginal returns to labor), possibly reflecting lack of competition, labor market regulations, or other non-productivity-based advantages.
- Interpretation of wedges:
  - Higher capital (labor) wedges indicate smaller frictions for capital (labor) to move into the sector (i.e., lower marginal returns).

### Cross-country heterogeneity
- Examples:
  - Italy: positive capital wedge in the business sector indicates over-accumulation of capital and low capital returns to the sector.
  - Portugal: low labor wedges in financial and ICTGS sectors indicate barriers to entry and frictions limiting employment in these sectors.
  - United States: misallocation driven by capital distortions, particularly in mining.
- Note: Pattern of factor input misallocation varies across countries beyond the common “favored” sectors.

### TFP gains from eliminating distortions (Table 1 summary)
- Full adjustment (Baseline Pareto) TFP level gains (Percent):
  - Austria: 9.9
  - Australia: 7.2
  - Denmark: 9.8
  - Finland: 10.9
  - Germany: 3.8
  - Italy: 19.5
  - Netherlands: 7.4
  - Portugal: 13.7
  - Sweden: 4.4
  - Spain: 6.2
  - United Kingdom: 5.3
  - Japan: 9.2
  - United States: 6.7
- With labor skill composition (Column 2; results reported as averages for 2000–07 or latest available; note for one column results are 2000–05 average):
  - Austria: 6.3
  - Australia: 3.7
  - Denmark: 5.1
  - Finland: 5.2
  - Germany: 1.9
  - Italy: 4.5
  - Netherlands: 5.3
  - Portugal: 12.0
  - Sweden: 2.1
  - Spain: 3.2
  - United Kingdom: 3.1
  - Japan: 4.5
  - United States: 3.2
- Decomposed contributions (selected entries, Percent):
  - Adjust Labor Allocation Only:
    - Austria: 3.3
    - Australia: 3.3
    - Denmark: 4.4
    - Finland: 5.4
    - Germany: 1.9
    - Italy: 14.3
    - Netherlands: 2.0
    - Portugal: 1.5
    - Sweden: 2.2
    - Spain: 2.9
    - United Kingdom: 2.1
    - Japan: 4.4
    - United States: 3.4
  - Adjust Capital Allocation Only:
    - Austria: 10.2
    - Australia: 8.7
    - Denmark: 10.7
    - Finland: 17.9
    - Germany: 8.1
    - Italy: 36.4
    - Netherlands: 12.1
    - Spain: 11.0
    - United Kingdom: 12.9
    - Japan: 13.7
    - United States: 6.2
    - (Table shows “-” for Portugal and Sweden in capital-only column)
- Key interpretation:
  - Average TFP gains across countries about 9 percent (Table 1 headline), ranging from 3.8 percent in Germany to 19.5 percent in Italy in the baseline full-adjustment scenario.
  - Accounting for labor skill composition tends to increase estimated TFP gains, indicating labor misallocation includes skill mismatches.
  - Relative contributions of labor vs. capital reallocation differ by country (e.g., labor adjustment more important in Austria and Portugal; capital adjustment more important in Italy).

### Growth implications from better input allocation
- Example projections assuming Pareto efficient allocation achieved within 10 years:
  - Italy: potential increase in annual growth rates of TFP (and GDP) by 1.8 percent per annum over the decade.
  - Portugal: potential increase by 1.3 percent per annum over the decade.
- Caveats:
  - Exercise assumes complete elimination of significant underlying distortions.
  - Estimates consider only cross-sector reallocation (not cross-industry or cross-firm), hence represent a lower bound of overall gains from reducing misallocation.

### Drivers of Total Factor Productivity growth
- Conceptual framework:
  - Uses “distance to the technology frontier” framework (Aghion and Howitt 2006, 2009; Acemoglu, Aghion, and Zilibotti 2006).
  - Distinguishes “imitation” (adoption of existing vintage technologies) from “innovation” (new state-of-the-art technology).
  - Countries farther from frontier grow mainly through adoption; those closer rely more on innovation.
- Structural drivers considered:
  - Product market regulation stringency (industry-specific OECD Indicators of Regulation Impact).
  - Employment protection rigidity (OECD Employment Protection Legislation indicator; economy-wide level).
  - Inputs supporting innovation and adoption: intensity of high-skilled labor, ICT capital inputs, R&D spending.

### Empirical findings on regulation, skills, ICT, and R&D (Figure 12 and Tables 2–3 summary)
- Regulatory and input patterns:
  - Services are among the most heavily regulated industries in most countries, especially Italy and Japan.
  - Wide dispersion in services regulation across peers indicates potential upside to reforms.
  - United States leads peers in most sectors in innovation indicators and intensive use of high-skilled labor.
  - In services, innovation often arises from tangibles (automation, ICT) and nontangibles (business processes), rather than R&D spending.
- Catch-up convergence and spillovers:
  - Productivity growth driven by catching-up (negative coefficient on TFP gap) and by developments at the frontier (positive coefficient on frontier TFP growth), consistent with knowledge spillovers.
  - Baseline results robust to alternative specifications and sample periods.

### Product market reforms: impacts and examples
- Finding:
  - Higher product market regulation in services is associated with lower TFP growth (Table 2).
  - Negative effect of stringent product market regulations on TFP growth is more severe the closer the industry is to the technology frontier.
  - Product market rigidity explains differences in ICT uptake and diffusion across economies.
- Policy implication:
  - Increasing competition in services by eliminating entry barriers, constraints to business operations, red tape and administrative burdens can provide productivity boosts to services sectors.
- Quantified example:
  - If Italy reduced its product market regulation to the level in the Netherlands (frontier in distribution services), the TFP growth gain within the services sector could amount to about 0.4 percent per year.

### Labor market rigidities
- Empirical summary:
  - Economy-wide employment protection indicator is not found to have significant explanatory power for industry-level TFP growth in the panel analysis, reflecting diversity of institutions and lagged effects of reforms.
  - Firm-level evidence suggests less stringent labor market institutions and more efficient human capital use (reducing skill mismatches) facilitate movement of labor to more productive firms and foster creative destruction.
  - Other literature finds mandatory dismissal regulations can depress productivity growth where layoff restrictions bind; longer unemployment benefits, high labor tax wedges, and certain collective bargaining systems can negatively affect employment.
- Policy implication:
  - Improving labor market institutions and reducing labor skill mismatches could enhance productivity by enabling labor reallocation to more productive firms and sectors.

*International Monetary Fund — Excerpt from chapter "Measuring misallocation across sectors" in THE NEW NORMAL*

### 37. Productivity impact of knowledge capital and innovation. More intensive use of high-

### 37. Productivity impact of knowledge capital and innovation. More intensive use of high-skilled labor and ICT capital inputs and higher spending on R&D activities is associated with higher productivity growth

### Core findings on knowledge capital, ICT, and R&D
- More intensive use of high-skilled labor, ICT capital inputs, and higher R&D spending is associated with higher productivity growth (Table 3).
- Channels differ:
  - High-skilled labor: fosters development of skill-intensive industries and diffusion of new technologies.
    - Example: increasing the share of high-skilled workers in the United Kingdom to the level of Japan could boost productivity growth by an additional 0.7 percent per year.
  - ICT diffusion: supports more efficient organizational structures and business processes within firms, yielding efficiency gains.
  - R&D spending: creates new efficiency-enhancing technologies.
- Policy implication: evidence supports a public policy role in improving educational attainment, fostering ICT diffusion, and encouraging innovation in advanced economies.

### Varying sectoral effects and spillovers
- Investments in human capital and R&D play a significant role in spurring TFP growth in ICT-related industries (and services in the case of skilled labor), with the strongest effects for countries closer to the frontier.
- Trade and technology spillovers imply reforms in leading economies generate important positive spillovers to other countries.
- Within services, the TFP impact of more intensive ICT use is highest for countries with lagging productivity in the sector, implying a role for fostering ICT investment and diffusion in services and eliminating barriers to trade and foreign direct investment.

### Quantified potential gains (bringing it together)
- Using TFP level estimates and improving each structural variable from the 25th percentile to the best-practice level, for the median country:
  - Improvements in structural variables could generate additional TFP growth gains of ¼ to ½ a percentage point annually across all industries.
- Table 4: TFP Growth Gains from Improvement in Structural Determinants (Percentage points, annually)
  - For each indicator, an “improvement” is defined as closing the gap between the highest-ranked country and the 25th percentile country.
  - Sector | Manufacturing | Services | ICT | All
    - Product Market Regulations | -0.1 | -0.1
    - Labor Skill | -1.4 | 1.2 | 0.4
    - ICT Capital | 0.7 | 0.7 | -0.3
    - R&D Expenditure | -- | 1.9 | 0.2
  - Interpretation: product market regulation gains accrue largely within services; R&D spending yields largest gains in ICT; better human capital and ICT use have broader productivity-enhancing effects.

### Identifying reform shocks and episodes
- Large reforms are identified as significant improvements of a structural indicator above a threshold.
  - For R&D, ICT capital, and high-skilled labor: a shock is identified if the change is above two standard deviations of the average annual change in the indicator.
  - For product and labor market regulations and infrastructure: a major reform is assumed when the change is below (above for infrastructure) one standard deviation.
- Added indicators for dynamic analysis: infrastructure (principal component of roads, phone lines, electricity generation capacity) and labor tax wedge (percentage of taxes and transfers paid in the share of total labor costs).
- Historical patterns:
  - Large infrastructure shocks mainly in the 1970s and have since waned.
  - Product market reforms spurted in the 1990s and continued into the 2000s.
  - Labor market reforms have been less frequent.
  - Positive ICT “shocks” increased since the 1970s but slowed on the eve of the crisis; human capital and R&D shocks also slowed.

### Dynamic effects: productivity over time
- Productivity impact typically grows over time; short- and medium-term effects are generally positive.
- Examples:
  - A 10 percent shock in R&D spending boosts TFP across all industries by 0.7 percent after three years and 2 percent over five years.
  - An ICT capital shock leads to a cumulative TFP level increase in ICT-related industries of about 30 percent after five years (i.e., two standard deviations of the average cumulative five-year change in TFP in the sample).
- Labor market nuance:
  - A reduction in employment protection is associated with a decline in TFP after three years, potentially reflecting weaker incentives for investment in firm-specific skills through excessive job turnover.
  - A reduction in the labor tax wedge is associated with higher productivity three years after the reform.
- Policy implication: different labor market reforms have heterogeneous efficiency impacts; avoid one-size-fits-all reforms and stay on policy “plateau” rather than cliffs.

### Heterogeneity across industries, initial settings, and frontier distance
- Effects are sector- and country-specific (distance to frontier matters).
  - Product market liberalization yields medium-term gains highest in ICT and personal services.
  - Lower tax wedges, higher R&D spending, and better human capital boost productivity in innovation-intensive manufacturing and ICT sectors.
  - Infrastructure has a positive productivity impact for all countries, irrespective of distance to the frontier.
- Conditional on initial settings:
  - Largest productivity gains from reducing product market regulations and labor tax wedges occur in highly regulated services sectors (not statistically significant for lightly regulated industries).
  - Medium-term impact of knowledge capital, innovation, and infrastructure is on average highest for industries with higher initial levels of innovation and infrastructure (e.g., manufacturing, ICT).
  - For ICT and personal services, the lower the initial levels of R&D spending and ICT capital use, the higher the productivity gains—underscoring the role of ICT diffusion where productivity lags.

### Reform impact across the business cycle
- Impact depends on overall economic conditions:
  - Infrastructure shocks are associated with significantly higher productivity gains during downturns.
  - Productivity gains from product market liberalization and innovation shocks tend to be higher during upturns.
- Policy implication: infrastructure investment can be particularly effective in spurring productivity in downturns; product market reforms and innovation may reap larger benefits in upturns, but liberalizing entry can still support investment and job creation during other phases.

### Dynamic effects on employment and output
- Short-term trade-offs:
  - Reforms could have small or negative short-term effects on output and employment due to reallocation costs, temporary unemployment, and fiscal/social costs.
  - Fiscal policy can facilitate reforms by bringing forward employment gains and mitigating short-term losses.
- Employment effects (cumulative three-year and five-year outcomes):
  - Product market liberalization and infrastructure shocks: positive and persistent impact on employment overall.
  - Intensive use of skilled labor: positive employment effects in manufacturing and ICT; lower employment in other production sectors (agriculture, utilities, mining).
  - Labor market liberalization (reduced dismissal costs): boosts employment in personal services in the short term.
  - ICT shocks: associated with lower employment in manufacturing and personal services (automation/displacement effects).
  - R&D shocks: associated with lower employment in distribution services.

*Source: IMF staff calculations and analysis as presented in the chapter "Productivity impact of knowledge capital and innovation" in the supplied content.*

### 49. Impact on output. Output gains from reforms are large and materialize predominantly in

### _sdn1503 - 49. Impact on output. Output gains from reforms are large and materialize predominantly in

### Impact on output: magnitudes and timing
- Output gains from reforms are large and materialize predominantly in the medium term (Figure 18).
- Product market liberalization and more intensive use of high-skilled labor raise output across all industries by about 2 percent over the medium term.
- Infrastructure shocks increase output across all industries by around 3 percent over the medium term.
- Innovation shocks (R&D/knowledge capital) increase output across all industries by around 5 percent over the medium term.
- Labor market liberalization, as captured by lower employment protection, has a negative aggregate short-term impact on output (though not on employment).
- Infrastructure shocks have a positive short- and medium-term output impact in almost all sectors, underscoring the role of infrastructure investment in supporting productivity, employment, and output.
- Note on Figure 18: The results show the cumulative three-year (ST) and five-year (MT) level gains in employment. Brown and shades of red indicate a negative statistically significant impact of the reform; shades of blue indicate a statistically significant positive one. “Other production” includes agriculture, forestry, fishing, mining, quarrying, electricity, gas, and water-related industries.

### Heterogeneity and interpretation: no one size fits all
- Productivity and growth effects of reforms depend on country-specific policy and institutional settings; reform priorities should be tailored to individual conditions.
- Empirical results highlight associations rather than reveal causation; they are illustrative of effective types of productivity-enhancing reforms, not prescriptions to be implemented identically by all countries.
- Product market rigidities vary across countries (e.g., outright entry barriers in business services in Europe; monopolistic structures in network industries).
- Labor market institutions vary considerably across advanced economies; reforms should account for different work models.
- Political economy considerations, social consequences of reforms, and implementation costs critically influence reform design and effectiveness.
- Complementarities (e.g., between product and labor market reforms) and sequencing of reforms across areas are material for realizing potential gains.
- Conclusion: there is neither a single reform path nor a silver bullet.

### Final remarks: priorities and policy implications
- Headwinds:
  - Secular headwinds for advanced economy policymakers arise from slowing pre-crisis trend growth in output and productivity.
  - These headwinds reflect structural changes as economies mature and relative factor costs adjust, alongside slowing innovation and human and physical capital accumulation, and policy distortions.
- Back to basics:
  - To increase growth potential, advanced economies must:
    - boost the productive capacity of their workforces;
    - make needed investments in physical and IT infrastructure;
    - encourage innovation;
    - remove binding product and labor market rigidities that impede resource allocation toward the most dynamic sectors and firms;
    - augment human capital, reduce skills gaps, and improve the quality of education.
  - The case for refocusing on these supply drivers is urgent given limits on policy support for demand and looming demographic headwinds.
- Investment in knowledge capital and innovation:
  - Innovation and investments in equipment, ICT, and new ideas are vital for raising long-term trend growth.
  - Investing in capital enables technology adoption and reorganization of production toward global best practices.
  - Competition and well-functioning markets help diffuse technology and reallocate resources from failing to growing firms.
  - Because market failures in innovation are common, the government has a role in providing incentives for R&D and broader supportive policies.
- Tailoring to economic conditions:
  - Where demand is weak, labor market slack exists, and investment efficiency is high, prioritize product market reforms and increased investment in productive infrastructure to reap immediate employment and output benefits.
  - Well-designed structural reforms that minimize potential short-run contractionary effects can translate into rising output and employment.

### Key takeaways and policy recommendations
- Services reform is a priority:
  - Services are the most important sector and the most highly regulated across advanced economies.
  - Significant upside scope exists from product market reforms to increase competition in services.
  - Investment in information technology can substantially boost service productivity in lagging regions.
  - Labor market policies should avoid excessive regulation and extreme disregard for labor conditions.
- Spillovers can be large:
  - Performance gains by technology leaders (e.g., Germany in manufacturing, the United States in ICT services) produce global spillovers that support productivity gains elsewhere.
  - Increasing returns to scale in knowledge capital create opportunities to boost domestic and global prospects, especially for countries far from the frontier.
- Infrastructure investment has clear benefits:
  - Investments in transport, energy, and telecommunications improve productivity, boost employment, and increase output.
  - These large-scale, long-term investments require high levels of coordination; governments often play a vital role in planning, delivering, and financing projects.
  - Debt-financed projects could have large productivity and output effects if clearly identified infrastructure needs are met through efficient investment (IMF 2014b).
- Improve resource allocation across sectors:
  - Reducing policy distortions that skew resource allocation toward low-value activities can yield significant productivity payoffs.
  - Rationalizing large agricultural subsidies and increasing competition in personal services can reduce misallocation.
- Calibrate reforms:
  - Reform design must be granular and country-specific; while some policies apply broadly, there is no one-size-fits-all strategy.
  - Divergence in productivity performance and drivers across sectors implies reform priorities must be appropriately calibrated.

### Annex: methodological note (Stochastic Frontier Analysis)
- Stochastic frontier analysis (SFA) is used to illustrate patterns of total factor productivity (TFP) growth across countries at the sector level.
- Concept: a world frontier technology exists for each industry/sector; country TFP depends on the world frontier TFP level and how efficiently the country uses that technology.
- Model specification close to Pitt and Lee (1981) with a world frontier productivity level determined by a constant and a trend component; TFP level estimates obtained by linking EU KLEMS TFP growth estimates with the Groningen Growth and Development Centre Productivity-Level database (PPP-adjusted industry TFP levels in 1997).
- Productivity level series constructed for 23 sectors spanning 1970–2007.
- Estimation uses the maximum likelihood method on 10-year rolling window sectoral TFP data; for each sector, the trend of frontier TFP growth and the average TFP level in percentage of world frontier are reported.

*International Monetary Fund — excerpt from the Staff Discussion Note chapter on "Impact on output" and related final remarks and annex.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/sdn/2015/_sdn1503.pdf_