## wpiea2020190-print-pdf - Section III uses the simple Aggregate Supply-Aggregate Demand (AS-AD) framework to

## Source details

**Canonical URL:** [wpiea2020190-print-pdf - Section III uses the simple Aggregate Supply-Aggregate Demand (AS-AD) framework to](https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020190-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2020/english/wpiea2020190-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2020/english/wpiea2020190-print-pdf.pdf.json)

---

### II. A FEW IMPORTANT MACROECONOMIC PUZZLES
- Documented facts:
  - Persistent low inflation since the global financial crisis despite closed output gaps and recovered unemployment in advanced economies and more recently in emerging markets.
  - A weakened unemployment–inflation link (flattening of the Phillips curve) since the 1990s; literature has not reached consensus on causes or future implications.
  - Rising inequality associated with suppressed wage growth.
- Hypothesized drivers of low inflation:
  - demographic changes;
  - technological changes (including ICT price declines and productivity effects);
  - globalization;
  - expansion of central bank independence and inflation targeting;
  - Fisherian low interest rate–low inflation relationships.
- Technological progress mechanisms lowering inflation:
  - suppressed unit labor costs and elevated productivity;
  - lower prices of ICT inputs reducing aggregate inflation weight;
  - heightened price competition and transparency;
  - suppression of wage growth.
- Summary: Low inflation, a weakened unemployment–inflation link, and rising inequality are well established, but integrating these structural changes into a simple macro framework across many economies remains challenging.

### III. CONCEPTUAL ADJUSTMENT TO THE AS-AD FRAMEWORK AND POLICY IMPLICATIONS
- Conceptual change:
  - Allow the short-term AS curve to become more elastic (flatten) while retaining the AS-AD framework.
  - Flattened short-term AS (AS → AS’) can reconcile closed output gaps (Y’’ ≈ Y*) with persistently lower price levels (P’’ < P*).
- Key policy implications (three main possibilities illustrated in Figures referenced in the source):
  1) Low inflation as a new long-term equilibrium:
     - E’’ at intersection of AD’’, AS’, and LS can imply a long-run price level P’’ below previous P*.
  2) Two-phase path to hit inflation target (expansion then stagnation):
     - Phase 1 (expansion): policy pushes growth above potential and raises inflation.
     - Phase 2 (stagnation): as near-term price rigidities wane and AS shifts (AS’ → AS’’), inflation may continue rising while growth slows.
     - This path differs from classical expectations where achieving an inflation target is accompanied by faster growth.
  3) Achieve inflation target without stagnation:
     - If potential growth rises to Y#, or AS flattening raises potential, policy moving AD to AD# can produce E# with inflation P* and higher growth Y#.
- Critical judgement:
  - Whether potential growth has changed over time matters for policy stance to avoid misjudgment consequences.

### IV. EMPIRICAL SUPPORT FOR MORE ELASTIC AGGREGATE SUPPLY
- Visual and econometric evidence:
  - Figure 5: Inflation gap versus output gap plots for the US and Germany show AS curves in the 1990s/early 2000s (red dashed) steeper than in later sample period (blue).
  - Regression and structural-break tests (Table 1):
    - In 16 out of 20 cases the AS slope was positively significant in the first subsample.
    - The slope became smaller in the latter period or turned negative (negative slopes were not significant).
    - Flattening was statistically significant in 14 cases.
    - Structural-break dates vary across countries; some coincide with the GFC (e.g., US, Australia) but most do not.
  - Robustness with import price controls (Table 2):
    - Controlling for import price inflation: out of 20 countries, 17 had positive and significant AS slope in the early subsample, and all 17 showed significant flattening in the latter period.
    - Alternative output gap measures (OECD vs WEO) and alternative inflation forecasts (Consensus long-term vs WEO 5-year) yield qualitatively similar results.
- Selected numeric examples from structural-break tests (Table 1):
  - USA: Before break 0.321***; After break 0.081**; P-value 0.00; Break point 2005q4; Sample start 1990q1; Sample end 2019q3.
  - GBR: Before break 0.441***; After break -0.52; P-value 0.00; Break point 2009q4; Sample 1990q1–2019q3.
  - DEU: Before break 0.211***; After break 0.061**; P-value 0.03; Break point 2002q3; Sample 1991q2–2019q3.
  - AUS: Before break 0.591***; After break -0.10; P-value 0.00; Break point 2008q3; Sample 1991q2–2019q3.
  - JPN: Before break -0.04; After break 0.00; P-value 0.82; Break point 2000q2; Sample 1990q1–2019q3.
- Interpretation: Empirical evidence supports a flatter short-term AS curve in many economies, motivating investigation of underlying drivers.

### V. INTANGIBLE INVESTMENT AS A POTENTIAL EXPLANATION
- Definition (from Haskel and Westlake (HW, 2018)):
  - Intangible investment examples: software; databases; research and development; exploration; creation of entertainment, literary or artistic originals; design; training; market research and branding; business process reengineering.
  - Tangible investment examples: ICT equipment (computer hardware), machinery and equipment.
- Four properties of intangible investment (HW):
  (a) Sunkenness: heavy sunk costs due to customization and lack of liquid secondary market.
  (b) Scalability: nonphysical assets can be employed in multiple places simultaneously.
  (c) Synergies: combining intangible and tangible inputs creates greater efficiencies or new products.
  (d) Spillovers: many intangible assets are non-rivalrous and non-excludable, generating external benefits.
- Mechanism linking intangibles to AS flattening:
  - Intangible capital implies high fixed (sunk) cost and low marginal cost.
  - Lower marginal cost reduces firms’ marginal cost and makes firm-level supply more price-sensitive; aggregate effect is a more elastic short-term AS.
- Thesis: The growing role of intangible investment helps explain how economies react to short-term demand and supply fluctuations and contributes to AS curve flattening and persistent low inflation.

### VI. DRIVERS IDENTIFIED FOR AS CURVE FLATTENING
- Two major drivers explored empirically:
  - (a) Globalization;
  - (b) The rising importance of intangible capital in production.
- These are long-term processes with heterogeneous timing and magnitudes across countries, consistent with non-synchronized structural breaks.

### EMPIRICAL FINDINGS ON TRADE OPENNESS AND INTANGIBLE CAPITAL (SAMPLE AND METHODOLOGY)
- Sample: all countries with available information from the Intangible Capital Database except Greece.
- Dependent variable: time series of AS curve slope estimated for each country using a 40-quarter rolling window.
- Two AS slope measures: with and without control for import price inflation.
- Robustness: results robust to excluding 2008–09 observations.

### COUNTRY-LEVEL COEFFICIENTS — TRADE OPENNESS (AS SLOPE WITHOUT IMPORT PRICE CONTROL)
- Coefficients (significance indicated exactly as reported):
  - AUT: -0.61***
  - BEL: -0.77***
  - DEU: -0.57***
  - DNK: -0.09
  - ESP: -1.03***
  - FIN: -0.31***
  - FRA: -0.25
  - GBR: -1.04***
  - ITA: -1.33***
  - NLD: -0.36***
  - PRT: -1.17***
  - SWE: -0.64***
  - USA: -2.36***
- Corresponding N and r2 examples:
  - AUT N: 76; r2: 0.35
  - GBR N: 76; r2: 0.16
  - NLD N: 76; r2: 0.86
  - USA N: 76; r2: 0.63

### COUNTRY-LEVEL COEFFICIENTS — TRADE OPENNESS (WITH IMPORT PRICE CONTROL)
- Coefficients:
  - AUT: -0.99***
  - BEL: -0.71***
  - DEU: -0.60***
  - DNK: -0.15
  - ESP: -1.11***
  - FIN: -0.41***
  - FRA: -0.36
  - GBR: -0.10
  - ITA: -1.60***
  - NLD: -0.47***
  - PRT: -1.30***
  - SWE: -1.11***
  - USA: -2.50***
- Corresponding N and r2 examples:
  - AUT N: 73; r2: 0.55
  - ITA N: 73; r2: 0.45
  - USA N: 73; r2: 0.55

### COUNTRY-LEVEL COEFFICIENTS — INTANGIBLE CAPITAL (AS SLOPE WITHOUT IMPORT PRICE CONTROL)
- Coefficients (intangible capital measured as ratio to stock of tangible capital):
  - AUT: -0.83***
  - BEL: -2.83***
  - DEU: -1.98***
  - DNK: -0.46**
  - ESP: -3.27***
  - FIN: -0.92***
  - FRA: -2.28***
  - GBR: -0.01
  - ITA: -4.21***
  - NLD: -0.93***
  - PRT: -2.75***
  - SWE: -0.64**
  - USA: -0.71***
- Corresponding N and r2 examples:
  - ESP N: 56; r2: 0.67
  - ITA N: 60; r2: 0.53
  - USA N: 64; r2: 0.72

### COUNTRY-LEVEL COEFFICIENTS — INTANGIBLE CAPITAL (WITH IMPORT PRICE CONTROL)
- Coefficients:
  - AUT: -1.21***
  - BEL: -2.60***
  - DEU: -1.93***
  - DNK: -0.70***
  - ESP: -3.39***
  - FIN: -1.11***
  - FRA: -2.59***
  - GBR: 3.09***
  - ITA: -4.89***
  - NLD: -1.17***
  - PRT: -3.16***
  - SWE: -1.59***
  - USA: -0.75***
- Corresponding N and r2 examples:
  - AUT N: 61; r2: 0.42
  - ITA N: 61; r2: 0.55
  - USA N: 61; r2: 0.53

### JOINT REGRESSIONS (TRADE OPENNESS AND INTANGIBLE CAPITAL) — SELECTED COUNTRIES (WITHOUT IMPORT PRICE CONTROL)
- Trade openness coefficients (selected):
  - AUT: -0.75***
  - BEL: -0.81***
  - DEU: -0.45***
  - DNK: 0.46***
  - ESP: -0.44**
  - FIN: -0.11*
  - FRA: -0.49*
  - GBR: -1.21***
  - ITA: -1.21***
  - NLD: -0.21***
  - PRT: -0.77***
  - SWE: -0.64***
  - USA: -1.01***
- Intangible capital coefficients (same specification, selected):
  - AUT: 0.06
  - BEL: -0.79
  - DEU: -0.41
  - DNK: -1.54***
  - ESP: -2.51***
  - FIN: -0.86***
  - FRA: -1.63***
  - GBR: 1.80**
  - ITA: -2.45***
  - NLD: -0.47***
  - PRT: -1.44***
  - SWE: -0.14
  - USA: -0.47***
- Corresponding N and r2 examples:
  - N: 64 for many country regressions; r2 examples include 0.43 (AUT), 0.90 (NLD), 0.76 (USA).

### CROSS-COUNTRY PANEL REGRESSIONS (TABLE 6) — AS SLOPES WITHOUT IMPORT PRICE CONTROL
- Trade openness coefficients across specifications (columns (1)–(6)):
  - -0.60***; -0.43***; -0.43***; -0.19***; -0.11**; -0.09*
- Intangible capital 1 coefficients where included:
  - -1.08***; -0.63***; -0.27***; -0.25**
- Intangible capital 2 (alternative measure) coefficients:
  - -0.51***; -0.31***; -0.37***; -0.36***
- Sample sizes and fit:
  - N (various specifications): 968; 812
  - r2 (various specifications): 0.32; 0.30; 0.31; 0.39; 0.40; 0.49; 0.51; 0.53; 0.51; 0.53
- Controls and estimation notes:
  - All regressions include country fixed effects; columns (6) to (10) also include time dummies.
  - Legend: * p<.1; ** p<.05; *** p<.01

### EMPIRICAL SUMMARY
- Trade openness:
  - In most countries, higher trade openness is associated with a flatter AS curve (negative and significant coefficients).
- Intangible capital:
  - In most countries, higher shares of intangible capital are associated with flatter AS curves (negative and significant coefficients), with the UK as a notable exception in some specifications.
- Joint inclusion:
  - When trade openness and intangible capital are included together, both tend to contribute to explaining a flatter AS curve though collinearity can reduce significance in some country regressions.
- Cross-country panel evidence:
  - Strong support that higher global integration and larger shares of intangible capital are associated with flatter AS curves.

### MECHANISMS: WHY GLOBALIZATION AND INTANGIBLES FLATTEN AS
- Common channel:
  - Both trends lower marginal cost of production, increasing producers' sensitivity to price changes and flattening the short-term AS curve.
- Globalization mechanisms:
  - Enables global sourcing of intermediate inputs.
  - Production fragmentation allows sourcing from cost-efficient suppliers, increasing competition and reducing marginal cost sensitivity to domestic adjustments.
- Intangible capital mechanisms:
  - Scalability: high sunk costs but low marginal cost of serving additional users or markets (examples: video streaming, pharmaceuticals, automobile driver assistance software).
- Evolution:
  - Globalization may have matured for the near term; intangible capital appears to be accelerating and spreading across sectors and borders.
  - Conventional products and services increasingly embed intangible components, implying continued growth in production processes that leverage intangible investment.

### SHORT-TERM MACRO AND POLICY-RELEVANT CONSEQUENCES
- Conceptual macro implication:
  - Rising intangible investment increases short-term aggregate supply elasticity (flatter short-term AS curve), reconciling closed output gaps with lower inflation.
  - The framework remains agnostic about long-term potential growth (vertical long-term AS curve unchanged in the conceptual setup).
- Policy-relevant implications and areas for further research:
  - Labor market and distributional effects:
    - Greater scalability and substitutability may reduce bargaining power for labor and lower wage responsiveness to output changes (flatten Phillips curve).
    - Possible larger income share accruing to capital; labor substitutable by intangible capital most at risk, while labor complementing intangible capital may benefit.
    - Potential exacerbation of inequality; implications for redistribution policies.
  - Market structure and competition:
    - Returns to scale may reinforce incumbents' advantages and increase market concentration.
    - Consumers may benefit from lower prices, but competition and innovation dynamics require further study at the sector level.
  - Financing intangible investment:
    - High fixed costs and ex ante risks favor internal financing or equity-type sources such as venture capital; bank-based financing may be reduced until market value is proven.
  - Taxation and fiscal policy:
    - If intangible-driven returns favor monopolistic competition, fiscal revenue strategies should be studied to support adversely affected workers.
    - Policy adaptation may be required to support transition and compensate those negatively impacted by changing competition and distribution dynamics.
- Broader conclusion:
  - The rising share of intangible investment has important implications for short-term fluctuations (notably persistent low inflation amid tight labor markets) and longer-term economic evolution.
  - The paper links intangible investment and globalization to a flatter short-term AS curve and persistent low inflation and provides a simple framework plus empirical evidence for these links.

*Source: wpiea2020190-print-pdf - Section III argues that considering supply-side changes makes the AS-AD framework more*

### Section III uses the simple Aggregate Supply-Aggregate Demand (AS-AD) framework to

### wpiea2020190-print-pdf - Section III uses the simple Aggregate Supply-Aggregate Demand (AS-AD) framework to

### II. A FEW IMPORTANT MACROECONOMIC PUZZLES
- Persistent low inflation since the global financial crisis, even as output gaps closed and unemployment recovered to record low levels prior to the Covid-19 downturn in 2020, is documented for advanced economies and more recently for emerging markets.3
- Hypothesized drivers of low inflation include:
  - demographic changes;
  - technological changes (including ICT price declines and productivity effects);
  - globalization;
  - expansion of central bank independence and inflation targeting;
  - Fisherian low interest rate–low inflation relationships.4
- A related puzzle is the apparent breakdown or “flattening” of the Phillips curve (weaker relationship between unemployment and inflation) since the 1990s; literature has not reached consensus on causes or future implications.5
- Technological progress is argued to reduce inflation through mechanisms such as:
  - suppressed unit labor costs and elevated productivity (Greenspan testimony, 2005);
  - lower prices of ICT inputs reducing aggregate inflation weight;
  - heightened price competition and transparency;
  - suppression of wage growth.6
- Suppressed wage growth has been associated with rising inequality and linked in empirical studies to technological change; other contributing explanations include weaker bargaining power of labor, automation, and country-specific institutional factors.7,8,9
- Summary conclusion: Low inflation, a weakened unemployment–inflation link, and rising inequality are well established, but integrating structural changes into a simple macro framework applicable across many economies remains challenging.10

### III. CAN THE AS-AD FRAMEWORK STILL EXPLAIN LOW INFLATION?
- Conceptual adjustment: Allow the short-term aggregate supply (AS) curve to become more elastic (flatten) relative to past, while keeping the AS-AD framework.
  - Standard equilibrium: E* at intersection of LS (vertical long-term supply indicating potential output Y*), downward-sloping AD, and upward-sloping short-term AS.
  - With a large negative AD shock economy moves to E’ (lower Y’, P’). Policy stimulus can move AD back toward AD’’ producing partial recovery E’’ in original formulation.
  - Puzzle: In many countries post-GFC, stimulus narrowed/closed output gaps (Y’’ ≈ Y*) but observed price level P’’ remained well below pre-crisis P*.
  - Reconciliation: If short-term AS flattens from AS to AS’, then even with output Y’’ = Y* the price level can remain at P’’ < P*.
- Key policy implications of a flattened AS curve:
  1) Low inflation could be a new long-term equilibrium.
     - In Figure 2, E’’ lies at intersection of AD’’, AS’, and LS, implying lower price level P’’ can be a long-run equilibrium rather than evidence of slack.
  2) Two-phase path to hit inflation target (Figure 3: expansion then stagnation):
     - Phase 1 (expansion): Supportive policy pushes growth above potential and raises inflation.
     - Phase 2 (stagnation): As near-term price rigidities wane and the AS curve shifts from AS’ to AS’’, inflation may continue to rise while growth slows.
     - This differs from classical Figure 1 expectation where reaching inflation target would be accompanied by faster growth.
  3) Achieve inflation target without stagnation (Figure 4):
     - If potential growth rises to Y#, or AS flattening raises potential, policy support moving AD to AD# can produce a new equilibrium E# with inflation P* and higher growth Y#.
- Critical policy judgement: Whether potential growth has changed over time matters for choosing policy stance and avoiding misjudgment consequences.

### IV. EMPIRICAL SUPPORT FOR MORE ELASTIC AGGREGATE SUPPLY
- Visual evidence:
  - Figure 5: Plots of inflation gap versus output gap for the US and Germany indicate AS curves in the 1990s/early 2000s (red dashed) were steeper than in later sample period (blue), i.e., flattened AS observed for most economies (exceptions include Japan and Mexico).
- Econometric/structural-break evidence (summary):
  - Regression and structural break tests (Table 1) show:
    - In most cases (16 out of 20), the slope of the AS curve was positively significant in the first subsample period.
    - The slope became smaller in the latter period or turned negative in some cases (none of the negative slopes were significant).
    - The flattening of the AS curve was statistically significant in 14 cases.
  - Structural break timings vary across countries (not synchronized); some coincide with the GFC (e.g., US, Australia) but most do not.
  - Robustness with import price controls (Table 2):
    - Controlling for import price inflation strengthens results: out of 20 countries, 17 (vs 16 in Table 1) had positive and significant AS slope in the early subsample, and all 17 showed significant flattening in the latter period (compared with 14 cases in Table 1).
    - Alternative output gap measures (OECD baseline vs WEO) and alternative inflation forecasts (Consensus long-term vs WEO 5-year) yield qualitatively similar pictures, though some country results are less robust.
- Selected numeric examples from structural break tests (Table 1 entries):
  - USA: Before break 0.321***; After break 0.081**; P-value 0.00; Break point 2005q4; Sample start 1990q1; Sample end 2019q3.
  - GBR: Before break 0.441***; After break -0.52; P-value 0.00; Break point 2009q4; Sample 1990q1–2019q3.
  - DEU: Before break 0.211***; After break 0.061**; P-value 0.03; Break point 2002q3; Sample 1991q2–2019q3.
  - AUS: Before break 0.591***; After break -0.10; P-value 0.00; Break point 2008q3; Sample 1991q2–2019q3.
  - JPN: Before break -0.04; After break 0.00; P-value 0.82; Break point 2000q2; Sample 1990q1–2019q3.
- Interpretation: The evidence supports the empirical relevance of a flattened AS curve in many economies, motivating investigation of underlying drivers.

### V. CAN INTANGIBLE INVESTMENT EXPLAIN THE AS CURVE FLATTENING?
- Definition: Intangible investment = nonphysical investment by firms or the state to produce products (examples from Haskel and Westlake (HW, 2018)): software; databases; research and development; exploration; creation of entertainment, literary or artistic originals; design; training; market research and branding; business process reengineering.
  - Tangible investment examples: ICT equipment (computer hardware), machinery and equipment.
  - Note: Overlap exists between “technological change” and intangible investment but they are not identical (e.g., software is intangible; hardware is tangible).
- Four key properties of intangible investment (HW):
  (a) Sunkenness:
    - Intangible investments often imply heavy sunk costs due to customization and lack of liquid secondary market (e.g., brand).
  (b) Scalability:
    - Nonphysical assets can be employed in multiple places simultaneously; intangible investment is more scalable than tangible investment.
  (c) Synergies:
    - Intangible investments combine with tangible and intangible inputs to produce greater efficiencies or new products (example: smartphones + apps → ride-sharing).
  (d) Spillovers:
    - Many intangible assets are non-rivalrous and non-excludable, generating external benefits (examples: design, training).
- Mechanism linking intangibles to a flatter AS:
  - Focus on scalability: Intangible capital’s marginal cost is lower than that for tangible capital—high fixed (sunk) cost but low marginal cost.
  - Marginal cost translates into a firm’s supply curve; firms with more intangible investment have lower marginal costs, so aggregate supply becomes more price-sensitive (more elastic) if intangible intensity rises across firms.
- Thesis: The growing role of intangible investment—including but not restricted to intangible aspects of technological innovation—is crucial to understanding how economies react to short-term demand and supply fluctuations and can help explain AS curve flattening and persistent low inflation.

### VI. WHAT HAVE BEEN DRIVING THE FLATTENING OF THE AS CURVE?
- The paper identifies two major drivers of AS curve flattening to be explored empirically in subsequent sections:
  - (a) Globalization;
  - (b) The rising importance of intangible capital in production.
- These are long-term processes that unfolded at different times and magnitudes across countries, consistent with the non-synchronized structural breaks in AS slopes observed empirically.

*Source: wpiea2020190-print-pdf - Section III uses the simple Aggregate Supply-Aggregate Demand (AS-AD) framework to*

### Section III argues that considering supply-side changes makes the AS-AD framework more

### wpiea2020190-print-pdf - Section III argues that considering supply-side changes makes the AS-AD framework more

### Empirical findings on AS curve flattening
- Sample: all countries with available information from the Intangible Capital Database with the exception of Greece.
- Dependent variable: time series of the AS curve slope estimated for each individual country using a 40-quarter rolling window.
- Two AS slope measures used: with and without control for import price inflation.
- Regressions tested robustness to exclusion of 2008-09 observations (results qualitatively the same or stronger).

- Trade openness (ratio of total exports and imports of goods and services to GDP) — country regressions (AS curve slope without control for import price inflation):
  - AUT: -0.61***
  - BEL: -0.77***
  - DEU: -0.57***
  - DNK: -0.09
  - ESP: -1.03***
  - FIN: -0.31***
  - FRA: -0.25
  - GBR: -1.04***
  - ITA: -1.33***
  - NLD: -0.36***
  - PRT: -1.17***
  - SWE: -0.64***
  - USA: -2.36***
  - Corresponding N: 76, 72, 72, 76, 68, 76, 76, 76, 76, 76, 72, 76, 76
  - Corresponding r2: 0.35, 0.49, 0.72, 0.02, 0.56, 0.14, 0.02, 0.16, 0.38, 0.86, 0.77, 0.26, 0.63

- Trade openness (AS curve slope with control for import price inflation):
  - AUT: -0.99***
  - BEL: -0.71***
  - DEU: -0.60***
  - DNK: -0.15
  - ESP: -1.11***
  - FIN: -0.41***
  - FRA: -0.36
  - GBR: -0.10
  - ITA: -1.60***
  - NLD: -0.47***
  - PRT: -1.30***
  - SWE: -1.11***
  - USA: -2.50***
  - Corresponding N: 73, 72, 69, 76, 68, 73, 73, 73, 73, 73, 72, 73, 73
  - Corresponding r2: 0.55, 0.35, 0.64, 0.03, 0.63, 0.23, 0.03, 0.00, 0.45, 0.90, 0.75, 0.60, 0.55

- Intangible capital (measured as ratio to stock of tangible capital) — AS curve slope without control for import price inflation:
  - AUT: -0.83***
  - BEL: -2.83***
  - DEU: -1.98***
  - DNK: -0.46**
  - ESP: -3.27***
  - FIN: -0.92***
  - FRA: -2.28***
  - GBR: -0.01
  - ITA: -4.21***
  - NLD: -0.93***
  - PRT: -2.75***
  - SWE: -0.64**
  - USA: -0.71***
  - Corresponding N: 64, 60, 60, 64, 56, 64, 64, 64, 60, 64, 64, 64, 64
  - Corresponding r2: 0.28, 0.50, 0.61, 0.08, 0.67, 0.68, 0.43, 0.00, 0.53, 0.80, 0.76, 0.10, 0.72

- Intangible capital (AS curve slope with control for import price inflation):
  - AUT: -1.21***
  - BEL: -2.60***
  - DEU: -1.93***
  - DNK: -0.70***
  - ESP: -3.39***
  - FIN: -1.11***
  - FRA: -2.59***
  - GBR: 3.09***
  - ITA: -4.89***
  - NLD: -1.17***
  - PRT: -3.16***
  - SWE: -1.59***
  - USA: -0.75***
  - Corresponding N: 61, 60, 57, 64, 56, 61, 61, 61, 61, 61, 61, 61, 61
  - Corresponding r2: 0.42, 0.34, 0.43, 0.11, 0.71, 0.74, 0.38, 0.19, 0.55, 0.83, 0.81, 0.39, 0.53

- Joint regressions (trade openness and intangible capital) — AS curve slope without control for import price inflation (selected country coefficients):
  - Trade openness:
    - AUT: -0.75***
    - BEL: -0.81***
    - DEU: -0.45***
    - DNK: 0.46***
    - ESP: -0.44**
    - FIN: -0.11*
    - FRA: -0.49*
    - GBR: -1.21***
    - ITA: -1.21***
    - NLD: -0.21***
    - PRT: -0.77***
    - SWE: -0.64***
    - USA: -1.01***
  - Intangible capital (same specification):
    - AUT: 0.06
    - BEL: -0.79
    - DEU: -0.41
    - DNK: -1.54***
    - ESP: -2.51***
    - FIN: -0.86***
    - FRA: -1.63***
    - GBR: 1.80**
    - ITA: -2.45***
    - NLD: -0.47***
    - PRT: -1.44***
    - SWE: -0.14
    - USA: -0.47***
  - Corresponding N: 64, 60, 60, 64, 56, 64, 64, 64, 60, 64, 64, 64, 64
  - Corresponding r2: 0.43, 0.62, 0.68, 0.26, 0.71, 0.69, 0.47, 0.10, 0.65, 0.90, 0.82, 0.28, 0.76

- Cross-country panel regressions (Table 6) — AS curve slopes estimated without control for import price inflation:
  - Trade openness coefficients across columns (1)–(6): -0.60***; -0.43***; -0.43***; -0.19***; -0.11**; -0.09*
  - Intangible capital 1 coefficients (where included): -1.08***; -0.63***; -0.27***; -0.25**
  - Intangible capital 2 coefficients (alternative measure): -0.51***; -0.31***; -0.37***; -0.36***
  - N (various specifications): 968, 812
  - r2 (various specifications): 0.32, 0.30, 0.31, 0.39, 0.40, 0.49, 0.51, 0.53, 0.51, 0.53
  - All regressions include country fixed effects; columns (6) to (10) also include time dummies.
  - Legend: * p<.1; ** p<.05; *** p<.01

- Empirical summary statements:
  - In most countries, higher trade openness is associated with a flatter AS curve (negative and significant coefficients).
  - In most countries, higher shares of intangible capital are associated with flatter AS curves (negative and significant coefficients), with the UK as a notable exception in some specifications.
  - When trade openness and intangible capital are included together, both tend to contribute to explaining a flatter AS curve though collinearity can reduce significance in some country regressions (e.g., Austria and Belgium).
  - Cross-country panel evidence strongly supports that higher global integration and larger shares of intangible capital are associated with flatter AS curves.

### Mechanisms: why globalization and intangible capital flatten AS
- Common channel: both trends lower marginal cost of production, increasing producers' sensitivity to price changes and flattening the AS curve.
- Globalization mechanism:
  - Allows producers to source intermediate inputs globally.
  - Production can be broken into parts and sourced from most cost-efficient suppliers, increasing competition and reducing marginal cost sensitivity to domestic adjustments.
- Intangible capital mechanism:
  - Higher scalability: once sunk costs (e.g., content production, server storage, patents, software) are incurred, marginal costs of serving additional users or markets are relatively low.
  - Examples: video streaming, pharmaceuticals, automobile driver assistance software.

- Evolution over time:
  - Globalization may have matured for the near term, but intangible capital appears to be accelerating and spreading across sectors and borders.
  - Conventional products and services are increasingly embedding intangible components, suggesting continued growth in production processes that leverage intangible investment.

### Short-term macroeconomic implications and policy-relevant consequences
- Conceptual implication:
  - Rising intangible investment increases short-term aggregate supply elasticity (flatter short-term AS curve) in AS-AD framework.
  - This helps reconcile simultaneous observations of a closed output gap and lower inflation.
  - The framework remains agnostic about impacts on long-term potential growth (vertical long-term AS curve).

- Implications discussed for future research and policy:
  - Labor market and distributional effects:
    - Greater scalability and substitutability may reduce bargaining power for labor and lower wage responsiveness to output changes (flatten Phillips curve).
    - Possible larger share of income accruing to capital; labor substitutable by intangible capital most at risk, while labor complementing intangible capital may benefit.
    - Potential exacerbation of inequality; may inform redistribution policies.
  - Market structure and competition:
    - Improving returns to scale may reinforce cost advantages of large incumbents, potentially increasing market concentration.
    - Consumers may benefit from lower prices, but competition and innovation could decline; sector-level implications merit study.
  - Financing of intangible investment:
    - Higher fixed costs and ex ante risks suggest intangible investment is better suited to internal financing or equity-type sources such as venture capital.
    - Role of bank-based financing may be reduced until investments prove market value.
  - Taxation and fiscal policy:
    - If intangible-driven returns favor monopolistic competition, fiscal revenue strategies should be studied to support workers adversely affected.
    - Policy adaptation may be required to support transition and compensate those negatively impacted by changing competition and distribution dynamics.

- Broader conclusion:
  - The rising share of intangible investment has important implications for short-term fluctuations (notably persistent low inflation amid tight labor markets) and longer-term economic evolution.
  - The paper provides a simple framework and empirical evidence linking intangible investment and globalization to a flatter short-term AS curve and persistent low inflation.

*Italicized source: IMF Working Paper (content unit: wpiea2020190-print-pdf - Section III argues that considering supply-side changes makes the AS-AD framework more)*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020190-print-pdf.pdf_