## 1. Italy: Gross Value Added Growth and Contributions 1/

## Source details

**Canonical URL:** [1. Italy: Gross Value Added Growth and Contributions 1/](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10244.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10244.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10244.pdf.json)

---

### Introduction and overall conclusion
- The financial crisis will likely have a long-lasting impact on Italy’s economic potential.
- Innovation and investment opportunities may weaken because demand prospects are likely to be poor and the real cost of borrowing remains high.
- Some increase in unemployment may be structural as displaced workers find it hard to return to the labor market amid industrial restructuring.
- Output is not expected to rebound to its precrisis trend over the medium term; structural weaknesses will continue to weigh on the Italian economy unless policy actions are taken.

### Precrisis growth and productivity (Italy’s Achilles’ heel)
- Real GDP growth averaged 1.6 percent during 1995‒2007.
- Italy’s per capita income (PPP) has declined relative to the euro area since the mid-1990s.
- Weakness driven largely by declining total factor productivity (TFP):
  - TFP contributions decreased substantially over 1995‒2005, pervasive across sectors and especially pronounced in manufacturing and non-tradable sectors.
- Labor contributions:
  - Contribution of labor growth positive in recent years; hours worked contribution increased significantly.
  - Labor participation accounted for almost half of annual GDP growth in 2001‒2007.
  - Employment contributed substantially; average hours worked marginally negative; strong contribution from immigration.
- Possible drivers of weak productivity and competitiveness:
  - Relatively high tax ratios.
  - Heavy regulatory burden in labor and product markets and bureaucratic red tape.
  - Large share of small and medium-size enterprises limiting scale and technology transfer.
  - Industrial specialization in products with relatively low value added; significant decline in world market share since mid-1990s.

### Crisis impact on level and productivity
- Output contractions:
  - 2008: output contracted by 1.3 percent.
  - 2009: output contracted by 5.0 percent.
- Historical severity:
  - First quarter of 2009: growth decline of 6 percent (year-on-year).
  - The recession was the worst since World War II; largest historical contraction of Italian exports since the 1930s.
- Recent productivity developments:
  - Productivity plummeted since crisis onset.
  - Unit labor costs have soared; profitability squeezed.
  - Capital deepening showed resilience; unemployment rose modestly due to part-time schemes and declining hours worked.
  - The drop in TFP growth over 2008‒09 offset most resilience in capital and employment.

### Identification methods to separate temporary vs permanent losses
- Key approaches used:
  - Univariate Hodrick-Prescott (HP) filter (noted limitations: end-point bias; ignores other information).
  - Two multivariate unobserved component models:
    - Multivariate filter (MV).
    - Production function approach (PFA).
  - Historical approach: evidence from previous international crises.
- MV filter features:
  - Incorporates relationships among GDP, unemployment, core inflation, capacity utilization.
  - Assumes relationships stable despite large shocks; provides counterfactual if crisis were a “normal” recession.
- PFA features:
  - Uses production-function relationships, capacity utilization, and unobserved stochastic components to decompose permanent and cyclical parts of factors.

### Multivariate (MV) filter results and projections
- Output gaps and timing:
  - MV estimates positive output gaps for 2005‒08.
  - MV estimates negative output gaps for 2009‒2011 smaller than some projections and projects closure by 2012.
  - Output gap estimated to have declined sharply in 2009.
- Long-run level loss projection:
  - Projected real GDP level suggests an output loss of about 14 percent relative to precrisis trend (1998‒2004) by 2015.
- Model forecast (MV):
  - The model forecasts a negative output gap of minus 1.3 percent for 2010, which gradually declines to minus 0.4 percent in 2011 before closing in 2012.
  - The economy is expected to converge to its steady state growth by 2012.
  - The 2-standard-deviation confidence band is about +/- 1 percent of the estimated potential growth for Italy; confidence bands widen to about 4 percentage points (-2 to +2 percent).
- Labor market and utilization dynamics (MV):
  - Inflation consistent with output gap dynamics: core inflation declined with negative gaps and rose with positive gaps.
  - Utilization gaps rebound faster; unemployment gap lags and closes by 2014.
  - NAIRU: model forecasts NAIRU to increase by only 0.2 during 2008‒10 due to temporary lay-off and work-reducing measures.

### Production Function Approach (PFA) findings and projections
- Drivers of potential growth variation:
  - Major source: changes in labor participation; bulk of permanent variation driven by labor trends.
  - Cyclical variations in real GDP mainly driven by TFP fluctuations; TFP is highly pro-cyclical.
- Long-run patterns:
  - Since mid-1990s, TFP growth declined from 1 percent to zero.
  - Potential growth rose from 0.7 percent at end of 1992‒93 recession to over 2 percent just before the current slowdown.
  - Capital deepening stable at around 1 percent.
  - Unemployment rate countercyclical: falls by about 0.04 percent as output rises 1 percent above potential.
  - Labor participation broadly acyclical.
- Volatility and levels:
  - Implied output gap estimates from PFA exhibit higher volatility than MV estimates.
  - PFA projected real GDP level suggests an output loss of about 11 percent relative to precrisis trend (1998‒2004) by 2015.
  - Output gap from PFA estimated to have troughed in 2009 at 2.6 percent (signed as presented in source).
  - Potential output growth likely dropped by 2.7 percent in 2009, expected to increase to 0.4 percent in 2010 and reach steady-state rate of 0.8 percent.
  - NAIRU estimated to rise from 7.2 in 2009 to 7.9 percent by 2014.

### Key short- and medium-term numeric forecasts (from PFA Table 3)
- Real GDP (annual percentage change): 2009 -5.2, 2010 0.9, 2011 1.6, 2012 1.4, 2013 1.2, 2014 1.1
- Potential GDP (annual percentage change): 2009 -2.7, 2010 0.4, 2011 0.8, 2012 0.8, 2013 0.8, 2014 0.8
- Output gap (percent of natural rate of unemployment): 2009 -2.6, 2010 -2, 2011 -1.2, 2012 -0.6, 2013 -0.3, 2014 -0.1
- Natural rate of unemployment: 2009 7.2, 2010 7.7, 2011 7.7, 2012 7.8, 2013 7.8, 2014 7.9
- Employment (annual percentage change): 2009 -0.8, 2010 -0.4, 2011 0.1, 2012 0, 2013 0, 2014 0
- Unemployment rate (percent): 2009 7.6, 2010 7.9, 2011 7.9, 2012 7.9, 2013 7.9, 2014 7.9
- Labor productivity (annual percentage change): 2009 -3.2, 2010 -1.8, 2011 0, 2012 0.2, 2013 0.3, 2014 0.4

### Historical evidence and medium-term loss estimates
- Applying IMF (2009) OLS coefficients yields an estimated medium-term output decline of about 15 percent relative to precrisis trend (1998‒2004).
- Summary of estimated output losses by 2015 relative to precrisis trend (1998‒2004):
  - Multivariate Filter: 14 percent
  - Production function framework: 11 percent
  - Evidence from previous episodes: 15 percent
- At end-2009, output losses associated with the crisis estimated at about 125 billion of 2000 euro (about 10 percent of precrisis 1998‒2004 real GDP).
- Three quarters of these losses are estimated to be related to shortfalls in potential output.
- Estimated output loss by 2015 relative to precrisis trend ranges between 11 to 15 percent using different methodologies.

### Policy-relevant implications and recommended priorities
- Structural weaknesses predating the crisis—especially weak TFP growth—are central to Italy’s poor medium-term outlook.
- Without policy actions addressing:
  - low productivity growth,
  - heavy regulatory burdens,
  - high tax ratios, and
  - firm-size constraints,
  structural weaknesses will continue to weigh on recovery and long-term potential.
- Fiscal implications:
  - The profile of potential output and the output gap implies that fiscal structural deficits are underestimated.
  - Stronger fiscal adjustment will be required than current projections entail; consolidation in the authorities’ latest Stability Program would still not be sufficient to ensure a sustained reduction in public debt.
  - With lower real GDP growth over the medium term than currently projected, a stronger, expenditure-based adjustment effort would be needed to put debt on a declining path.
  - A more front-loaded fiscal adjustment would help balance intergenerational distribution of long-term fiscal adjustment arising from current pension reform design.
- Policy measures to limit permanent damage and shape medium-term dynamics:
  - Use of the wage supplementation fund (Cassa Integrazione Guadagni) with on-the-job training to cushion structural unemployment.
  - Apply the Lisbon Assessment Framework (LAF) to identify priorities to strengthen medium-term TFP growth: R&D and innovation, education, product and capital market regulation, and labor market policies to increase working time and make work pay.
- If structural reforms fail to restore growth, sizeable and more ambitious fiscal consolidation starting now will be required.

*Italic source attribution: IMF Working Paper chapter titled "1. Italy: Gross Value Added Growth and Contributions 1/" (extracted content).*

### 1. Italy: Gross Value Added Growth and Contributions 1/  ......................................................5

### 1. Italy: Gross Value Added Growth and Contributions 1/

### Introduction
- The financial crisis will likely have a long-lasting impact on Italy’s economic potential.
- Innovation and investment opportunities may weaken because demand prospects are likely to be poor and the real cost of borrowing remains high.
- Some increase in unemployment may be structural as displaced workers find it hard to return to the labor market amid industrial restructuring.
- The paper assesses Italy’s medium-term output losses following the crisis and implications for longer-term growth and the fiscal situation.
- Conclusion preview: output is not expected to rebound to its precrisis trend over the medium term; structural weaknesses will continue to weigh on the Italian economy unless policy actions are taken.

### Productivity: Italy’s Achilles’ Heel
- Italy experienced chronically low economic growth even before the global financial crisis:
  - Real GDP growth averaged 1.6 percent during the period 1995‒2007, down from over 2 percent in the earlier decade.
- Italy’s per capita income (measured in purchasing power parity) has declined, diverging away from the euro area over the same period.
- The dismal growth performance is largely due to poor productivity:
  - Breaking down GDP growth into labor, capital, and total factor productivity (TFP) shows anemic growth mostly explained by declining TFP.
  - TFP contributions decreased substantially over the period 1995‒2005—pervasive across all sectors but especially pronounced in manufacturing and non-tradable sectors.
- Labor contribution contrasted with TFP:
  - Contribution of labor growth has been positive over recent years.
  - Contribution of hours worked increased significantly—also relatively to the EU15—thanks to extensive labor market reforms.
  - Within labor, labor participation accounted for almost half of annual GDP growth in 2001‒2007.
  - Contribution of employment was substantial; average hours worked marginally negative; strong contribution from immigration.
- Possible drivers of weak productivity and competitiveness:
  - Relatively high tax ratios, deemed to have undercut growth by discouraging labor supply and investment.
  - Heavy regulatory burden in labor and product markets and bureaucratic red tape, likely hampering competition and stifling investment incentives.
  - Large share of small and medium-size enterprises, possibly limiting economies of scale and technology transfers.
  - Industrial specialization in products with relatively low value added contributed to steady erosion of competitiveness and a significant decline in Italy’s world market share in world trade since the mid-1990s.

### The Crisis: A New Toll on Productivity
- The global financial crisis exacerbated Italy’s structural weaknesses:
  - Output contracted by 1.3 percent in 2008 and 5.0 percent in 2009.
  - The downturn started earlier and has been deeper and longer-lasting than in most euro area peers.
  - The recession led to a sharp fall in exports; investment dropped more sharply than in earlier recessions; inventories were cut; private consumption declined significantly despite strong household balance sheets.
- Historical severity:
  - The economy suffered the worst recession since World War II.
  - In the first quarter of 2009, growth witnessed a decline of 6 percent (year-on-year), four times as large as during the EMS crisis.
  - Following the EMS crisis, output did not recover to its precrisis trend (1983‒89), resulting in permanent loss in potential output growth in the long run.
  - The recession featured the largest historical contraction of Italian exports since the 1930s.
- Recent productivity developments:
  - Since the onset of the crisis, productivity has plummeted further.
  - Unit labor costs have soared and profitability has been further squeezed, worsening Italy’s competitive position.
  - Capital deepening has shown strong resilience thus far, while unemployment has been rising only modestly, largely due to part-time work schemes and declining hours worked.
  - The drop in TFP growth over 2008‒09 has been so large it has offset most of the resilience in capital and—to a lesser extent—employment.

### Alternative Ways to Disentangle Temporary from Permanent Losses
- Key identification problem: difficult to distinguish decline in output due to persistent demand shock (temporary) versus supply factors (permanent).
- The crisis may have broken down previous economic relationships; structural changes can contribute substantially to output movements.
- Survey evidence:
  - Capacity utilization and expected capacity constraints indicate adverse demand shocks started in late 2008.
  - Financial conditions tightened before the collapse in capacity utilization; during 2009 demand collapsed and limited production.
- Approaches used to assess impact on potential output:
  - Statistical approaches:
    - Univariate Hodrick-Prescott (HP) filter.
    - Two multivariate unobserved component models:
      - a multivariate filter (MV), and
      - a production function approach (PFA).
  - Historical approach:
    - Evidence from previous international crises is considered to provide a more judgmental perspective.
- Comments on the HP filter:
  - HP filter uses only the series’ own data; ignores other relevant economic information.
  - Prone to “end-point bias,” especially problematic when prolonged recession or structural break occurs at the end of the sample.
  - HP-based estimates point to a pre-existing weakness in labor productivity trend growth, but trend measures based on HP-filtering are unreliable at the end points in the presence of a crisis.

### Multivariate (MV) Filter and Production Function Approach (PFA)
- Multivariate (MV) filter:
  - Incorporates both recent data and long-term trends.
  - Uses a small macroeconomic model estimating relationships between actual and potential GDP, unemployment, core inflation, and capacity utilization in manufacturing.
  - Assumes relationships among major economic variables were stable despite large shocks.
  - Provides the counterfactual dynamics if this were a “normal” recession (large shock but no structural break).
- MV filter results:
  - Estimates positive output gaps for 2005‒08, contrasting with IMF historical estimates that suggest potential output levels were overestimated.
  - Estimates negative output gaps for 2009‒2011 smaller than current projections; projects closure by 2012.
  - The projected real GDP level suggests an output loss of about 14 percent relative to precrisis trend (1998‒2004) by 2015.
  - Output gap is estimated to have declined sharply in 2009; Italy’s output gap troughed at about minus   2 percent in 1993 during the EMS crisis (comparison point).
- Production Function Approach (PFA):
  - Uses productivity-capacity utilization relations and unobserved stochastic components (analytical underpinnings reported in Appendix III).
  - Offers an alternative multivariate unobserved component model leveraging production-function relationships (details and numerical projections appear elsewhere in the chapter).

### Historical Evidence and Crisis Comparisons
- Comparative recession dynamics:
  - The 2008 recession was deeper than 1974‒75 oil-price crisis and the 1992‒93 EMS crisis.
  - Trough comparisons indicate larger and more synchronized declines in exports and investment in 2008 compared with previous episodes.
- Contributions to growth in times of crisis (high-level patterns, as reported):
  - During the financial crisis, TFP and labor productivity fell sharply; capital deepening showed resilience in the short run.
  - Table summaries indicate pronounced negative contributions from TFP in the financial crisis relative to prior crises.

### Policy-Relevant Implications (from the analysis)
- Structural weaknesses that predated the crisis—especially persistently weak TFP growth—are central to Italy’s poor medium-term outlook.
- Without policy actions to address:
  - low productivity growth,
  - heavy regulatory burdens,
  - high tax ratios, and
  - constraints associated with firm-size structure,
  structural weaknesses will continue to weigh on recovery and long-term potential.
- The paper implies the need for reforms to restore competitiveness and productivity growth so that output can converge back toward higher potential paths (specific policy prescriptions are discussed in later sections of the chapter).

*Italic source attribution: IMF Working Paper chapter titled "1. Italy: Gross Value Added Growth and Contributions 1/" (extracted content).*

### 2.5 percent. The model forecasts a negative output gap of minus 1.3 percent for 2010, which

### _wp10244 - 2.5 percent. The model forecasts a negative output gap of minus 1.3 percent for 2010, which

### Key short- and medium-term forecasts
- The model forecasts a negative output gap of minus 1.3 percent for 2010, which gradually declines to minus 0.4 percent in 2011 before closing in 2012.
- The economy is expected to converge to its steady state growth by 2012.
- The 2-standard-deviation confidence band is about +/- 1 percent of the estimated potential growth for Italy.
- There is a high degree of uncertainty around the forecast, with confidence bands widening to about 4 percentage points (-2 to +2 percent).

### Inflation, utilization, and labor market dynamics
- Inflation behavior is consistent with the model’s output gap dynamics: Italy’s core inflation declined during periods with negative output gap, and rose during years with positive gap.
- The recoveries in output and utilization gaps are expected to move in tandem, while the unemployment gap lags behind.
- The unemployment gap is influenced by the current and lagged output gap but has smaller cyclical fluctuations due to labor hoarding and the “discouraged worker” effect during recessions.
- The utilization gap exhibits more volatility with sharp declines during recessions; the utilization gap declined to in 1993 and to over 8 percent in 2009.
- Following the 2009 trough, capacity utilization and the output gaps rebound, closing by 2012; the unemployment gap closes by 2014.

### NAIRU and unemployment
- The NAIRU is expected to rise moderately. The estimated NAIRU peaked in 1998, then gradually declined before climbing up toward the end of 2009.
- The decline during 2008 and beginning of 2009 is likely due to the discouraged worker effect and falling participation rates.
- The model forecasts the NAIRU for Italy to increase by only 0.2 during 2008‒10, reflecting measures introduced in Italy for temporary lay-off and work reducing measures.
- The unemployment gap closes by 2014, reflecting persistence in the labor market. The NAIRU’s return to the steady state rate is slow.

### Production Function Approach (PFA) findings
- The PFA shows the major source of potential growth variation is associated with changes in labor participation; the bulk of permanent variation in output is driven by shifts in labor trends (labor participation and employment).
- Cyclical variations in real GDP are mainly driven by total factor productivity (TFP) fluctuations.
- TFP is highly pro-cyclical; its structural component diverges from potential output dynamics.
- Since the mid-1990s, TFP growth declined from one percent to zero.
- Potential growth rose from an annual rate of 0.7 percent at the end of the 1992‒93 recession to over 2 percent just before the current slowdown.
- The rate of capital deepening has remained stable over time, at around 1 percent.
- TFP and hours worked are strongly pro-cyclical; both plunged below trend since 2002 and became more pro-cyclical since 1999.
- The unemployment rate is significantly countercyclical: it falls by about 0.04 percent as output rises 1 percent above potential.
- Labor participation is broadly acyclical; there is positive comovement between average hours worked per employee, output, and productivity once structural shifts are identified.
- Implied output gap estimates from the PFA tend to exhibit higher volatility than estimates from the multivariate (MV) approach.

### PFA numerical projections and table-based indicators
- The projected real GDP level suggests an output loss of about 11 percent relative to precrisis trend (1998‒2004) by 2015 (alternative estimates reported elsewhere in the chapter).
- Output gap derived from the PFA is estimated to have troughed in 2009 at 2.6 percent.
- Potential output growth is likely to have dropped by 2.7 percent in 2009, is expected to increase to 0.4 percent in 2010 before reaching its steady-state rate of 0.8 percent.
- The NAIRU is estimated to rise gradually, from 7.2 in 2009 to 7.9 percent by 2014, when the unemployment gap is also expected to be reabsorbed.

- From Table 3 (annual percentage change unless noted otherwise):
  - Real GDP: 2009 -5.2, 2010 0.9, 2011 1.6, 2012 1.4, 2013 1.2, 2014 1.1
  - Potential GDP: 2009 -2.7, 2010 0.4, 2011 0.8, 2012 0.8, 2013 0.8, 2014 0.8
  - Output gap (percent of natural rate of unemployment): 2009 -2.6, 2010 -2, 2011 -1.2, 2012 -0.6, 2013 -0.3, 2014 -0.1
  - Natural rate of unemployment: 2009 7.2, 2010 7.7, 2011 7.7, 2012 7.8, 2013 7.8, 2014 7.9
  - Employment: 2009 -0.8, 2010 -0.4, 2011 0.1, 2012 0, 2013 0, 2014 0
  - Unemployment rate (percent): 2009 7.6, 2010 7.9, 2011 7.9, 2012 7.9, 2013 7.9, 2014 7.9
  - Labor productivity: 2009 -3.2, 2010 -1.8, 2011 0, 2012 0.2, 2013 0.3, 2014 0.4

### Evidence from previous international episodes and estimated medium-term losses
- Applying IMF (2009) OLS coefficients, the medium-term output is estimated to decline by about 15 percent relative to the precrisis trend (a result driven by a high precrisis investment share of GDP and Italy’s large initial output loss during the crisis).
- The chapter’s summary of estimated output losses relative to precrisis trend (1998‒2004):
  - Multivariate Filter: 14 (percent by 2015)
  - Production function framework: 11 (percent by 2015)
  - Evidence from previous episodes: 15 (percent by 2015)
- With output back to its end-2001 level, Italy’s output losses associated with the crisis at the end of 2009 are estimated to be about 125 billion of 2000 euro (about 10 percent of precrisis 1998‒2004 real GDP).
- Three quarters of these losses are estimated to be related to shortfalls in potential output.
- The estimated output loss by 2015 relative to precrisis trend (1998‒2004) ranges between 11 to 15 percent using different methodologies.

### Policy implications and recommended priorities
- A significant permanent output loss will likely be the legacy of the global financial crisis. The path of output level is not expected to rebound to its precrisis trend over the medium term, even though growth is projected to return to trend by 2012.
- In the short run, the decline in output growth is mainly accounted for by a collapse in productivity growth. Over the medium term, productivity is likely to recover and contribute to potential output growth by approximately 0.5 percent, while employment is expected to suffer more enduring losses.
- Capital accumulation is expected to remain weak over 2010 and, in the medium term, to contribute slightly less to growth than before the crisis.
- Stronger fiscal adjustment will be required:
  - The profile of potential output and the output gap implies that fiscal structural deficits are underestimated.
  - There will be a need for a stronger adjustment effort than current projections entail, and for reforms to stimulate faster growth.
  - With the forecasted real GDP growth, the consolidation envisaged in the authorities’ latest Stability Program would still not be sufficient to ensure a sustained reduction in public debt.
  - With lower real GDP growth over the medium term than currently projected, a stronger, expenditure-based, adjustment effort would be needed to put debt on a declining path.
  - A more front-loaded fiscal adjustment would also help balance, to some extent, the highly unequal intergenerational distribution of the long-term fiscal adjustment arising particularly from the current design of the pension reform.
- Policy can limit the damage by shaping medium-term dynamics and reducing permanent costs:
  - The wage supplementation fund (Cassa Integrazione Guadagni) involves on-the-job training which could cushion the impact of the crisis on structural unemployment.
  - Applying the Lisbon Assessment Framework (LAF) may help identify policy priorities to strengthen medium-term TFP growth: R&D and innovation, education, product and capital market regulation, and labor market policies to increase working time and make work pay.
- If growth cannot be resumed through structural reforms, sizeable fiscal adjustment will be required; this calls for a more ambitious fiscal consolidation starting now.

*Source: IMF staff calculations.*

### REFERENCES

### _wp10244 - REFERENCES

### References
- Cited works include empirical and methodological studies on potential output, HP filter issues, multivariate filtering, production-function decompositions, financial crises effects on potential output, and Italy-specific analyses. Authors and outlets include Bardone and Reitano; Bassanetti, di Antonio, Cecioni, Nobili, and Zevi; Benes et al.; Cerra and Saxena; Codogno and Felici; Cotis, Elmeskov, and Mourougane; Daveri and Jona-Lasinio; Dew-Becker and Gordon; European Commission; Furceri and Mourougane; Haugh, Ollivaud, and Turner; International Monetary Fund; and Sgherri.

### Appendix I — Features and Pitfalls of the HP Filter
- HP filter definition: minimizes sum of squared deviations of log variable y from trend τ with smoothness penalty parameter λ.
- Key properties and caveats:
  - Trend τ is a function of λ and both past and future values of y.
  - Higher λ → greater smoothness; very large λ → linear time trend; λ → 0 → trend coincides with series.
  - Standard λ recommendation: 100*s^2, where s denotes series frequency.
- Two major limitations highlighted:
  - End-of-sample problem: in-sample phase shift due to reliance on future information. Remedies include extending data out-of-sample via historical growth or AR forecasts, but these can bias end-of-series estimates if past growth is a poor proxy.
  - Mis-specification when cyclical fluctuations are highly persistent or when trend shocks have greater variance than cycle shocks; choice of λ implicitly assumes trend fluctuations account for 2½ percent of cyclical fluctuations in quarterly data (or 1 percent in annual data).
- Conclusion: HP trends can be misleading for analyzing ongoing prolonged slowdowns.

### Appendix II — A Multivariate Filter (MV Model)
- Model variables and definitions:
  - Output gap y_t: log difference between actual GDP (Y_t) and potential GDP (Y^p_t); approximately percent of potential output.
  - Unemployment gap u_t: difference between NAIRU (U^p_t) and actual unemployment U_t. A positive unemployment gap indicates excess demand for labor.
  - Capacity utilization gap c_t: difference between actual manufacturing capacity utilization index (C_t) and its equilibrium level (C^p_t).
  - Model focuses on core inflation to capture excess-demand–inflation relationship.
- Inflation equation: current core inflation affected by level y_t and change (y_t − y_{t−1}) of the output gap; previous period inflation coefficient set to one.
- Unemployment dynamics: Equation (5) links unemployment gap to output gap and lagged unemployment gap (Okun's law + lag structure).
- Capacity utilization dynamics: Equation (6) links capacity utilization gap to its lag and output gap.
- NAIRU dynamics: Equation (7) NAIRU influenced by its lag, transitory shocks, persistent shocks, the output gap, and deviation from long-run equilibrium. Persistent shocks follow AR process (equation (8)).
- Potential output equation: Equation (9) includes first difference of NAIRU with coefficient set to labor share in Cobb-Douglas (θ). Long-run difference (19 quarters) of NAIRU constrained to (1−θ).
- Equilibrium capacity utilization follows stochastic process with transitory and persistent shocks.
- Long-term inflation objective formulation considers revisions to previous expectations; historical expectations from Consensus Economics.
- Output gap influenced by monetary policy and stochastic term (equation (14)).
- Estimation approach and assumptions:
  - Bayesian estimation.
  - Sample period: 1992Q4 to 2009Q3.
  - Steady-state assumptions: labor share = 0.61; output growth = 0.7 percent; unemployment rate = 8.3 percent.
  - Results described as relatively robust; limited sensitivity of current quarter estimates to new data revisions.
- Data sources listed:
  - Y Gross Domestic Product (SAAR, Bil.Chn.2000.Euros)
  - C Capacity utilization in manufacturing sector (Haver)
  - Annual rate of core inflation (Haver)
  - Long term inflation expectations (Consensus Economics)
  - U Unemployment rate (SA, percent)

- Table 1 — Prior and Posterior (Maximum Regularised Likelihood)
  - Parameters with Prior Mode, Prior Dispersion, Posterior Mode, Posterior Dispersion:
    - alpha: 0.500, 0.016, 0.496, 0.024
    - beta: 0.400, 0.032, 0.218, 0.040
    - omega: 0.500, 0.032, 0.392, 0.045
    - rho1: 0.800, 0.016, 0.806, 0.024
    - kappa1: 0.100, 0.063, 0.427, 0.059
    - phi1: 0.800, 0.016, 0.813, 0.025
    - phi2: 0.300, 0.016, 0.252, 0.024
    - tau: 0.100, 0.016, 0.113, 0.022
    - delta: 0.500, 0.016, 0.498, 0.024
    - kappa2: 1.500, 0.158, 1.735, 0.141
    - parhist: 5.000, 0.316, 4.925, 0.472
    - rho2: 5.000, 0.316, 5.034, 0.468
    - lambda1: 1.000, 0.316, 1.061, 0.441
    - std_RES_Y: 1.000, 0.032, 0.906, 0.049
    - std_RES_G: 1.000, 0.032, 1.038, 0.049
    - std_RES_UNR_GAP: 0.500, 0.032, 0.330, 0.044
    - std_RES_UNR_BAR: 0.100, 0.016, 0.099, 0.024
    - std_RES_UNR_G: 0.100, 0.016, 0.117, 0.021
    - std_RES_CAPU_GAP: 0.400, 0.032, 0.569, 0.040
    - std_RES_CAPU_BAR: 0.250, 0.016, 0.274, 0.025
    - std_RES_CU_G: 0.075, 0.003, 0.076, 0.005
    - std_RES_PIE: 0.500, 0.032, 0.478, 0.039
    - std_RES_PIELTE: 0.300, 0.032, 0.159, 0.019

- Table 2 — Forecasting Accuracy and Revision Robustness
  - Mean absolute revisions (according to most recent estimates) — horizons: 1Q Ahead, 4Q Ahead, 8Q Ahead, 12Q Ahead:
    - LGDP: 0.598, 1.929, 2.882, 3.259
    - PIE: 40.273, 0.656, 0.826, 0.907
    - UNR: 0.228, 0.691, 1.164, 1.421
    - CAPU: 1.043, 2.717, 3.221, 3.067
    - PIELTE: 0.142, 0.326, 0.453, 0.621
  - Mean absolute revisions by vintage (quarter t-12, t-8, t-4, nowcast):
    - Y: 0.144, 0.176, 0.227, 0.295
    - Y (HP): 0.120, 0.177, 0.327, 0.498
    - UNR_GAP: 0.131, 0.125, 0.110, 0.096
    - UNR_GAP (HP): 0.076, 0.070, 0.177, 0.332
  - Root Mean Squared Errors reported (table heading present).

### Appendix III — Production Function with Unobserved Stochastic Components
- Production function specification (variable capital utilization, Cobb-Douglas with constant returns):
  - Y_t = A_t C_t^(1−β) K_t^β L_t^(1−β)  (equation (15) in log-linear form as equation (16): y_t = a_t + (1−β) c_t + β k_t + (1−β) l_t)
- Definitions and decomposition:
  - β is labor share (measured by cost of labor services as share of total costs).
  - A is total factor productivity (TFP); L total hours worked; K capital stock; C unobserved capacity utilization in (0,1].
  - All factor inputs decomposed into permanent (*) and cyclical (c) components, except capital stock assumed fully permanent.
  - Solow residual: permanent component *a driven by technology; cyclical component c_a absorbs non-technological effects and intensity of capital use.
- Labor (l) decomposed into working-age population (wpop), participation ratio (pr), unemployment rate (u), average hours (h); each split into permanent and cyclical parts.
- Multivariate unobserved-components (UC) model:
  - Measurement equation for real output (equation (19)) expresses output as function of permanent and transitory components.
  - Transition equations (20) specify dynamics: trends follow random walks with stochastic drifts; transitory components driven by a common second-order AR reference cycle Lφ constrained common across factor inputs and driven by industrial production index ip.
  - Transitory components vector ψ expressed as linear combinations of current and lagged values of reference cycle via loading matrix τ.
- Identification and estimation:
  - Relative importance of supply vs demand shocks determines smoothness of trend (ratio of variance of cycle to variance of trend).
  - Kalman filter and smoothing used for state extraction and maximum-likelihood estimation in state-space form.
  - Iteration strategy: basic filter from t=1 to T, evaluate log-likelihood from observation τ+1 to T to minimize initial-value effects; last basic filter iteration provides smoothing initial values.
  - Estimation implemented in Gauss 6.0.
- Notes and modeling choices:
  - Employment rate approximated by first-order Taylor: ln(1−u_t) ≈ −u_t to maintain log-linearity while modeling NAIRU.
  - Demand and supply shocks assumed orthogonal.
  - Reference to Sgherri (2004) for related exposition.

*Italic: Source: _wp10244 - REFERENCES (PDF chapter/section)._

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp10244.pdf_