Abstract
This Selected Issues paper examines external developments and competitiveness in France. Over the past decade, the current account has deteriorated from a surplus of 1.2 percent of GDP in 2002 to a deficit of about 2.3 percent in 2012, as France lost ground in goods trade and services recorded just a slight increase in global market shares. The slight improvement of the trade deficit seen in 2012 may suggest a change in trend, although it is still too early to determine. Past deterioration in export performance points to competitiveness weaknesses, rooted in significant rigidities in labor and product markets.
Potential GDP Estimates for France: Prudent (and Calling for Action)1
A. Introduction
1. The Fund has been recommending that countries adopt a structural approach to fiscal policy. The recently adopted EU Fiscal Compact requires countries to incorporate structural targets as part of their fiscal rules, against which fiscal performance will be evaluated by the European Commission (EC). However, implementation of structural targets rests on accurate estimation of potential output and potential growth.
2. Estimates can vary considerably depending on assumptions and estimation techniques. Every method requires a number of arbitrary choices, either at the level of parameters (in statistical techniques) or the underlying economic theory (in structural approaches). As a result, real time estimates of potential growth are often subject to significant revisions. By way of illustration, in hindsight, the EC and the Fund would have overestimated France potential GDP growth over 2005-11 by roughly the same amount—0.2 percentage points on average. The largest downward revisions (of around 0.6-0.7 percentage points) were recorded in 2007 and 2011.
France - Revisions in GDP Potential Growth
(2005-2011)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: AMECO and WEO databases; and Staff calculations.Note: The real-time and revised estimates are those produced in the Spring of years t and t+2 respectively.France - Revisions in GDP Potential Growth
(2005-2011)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: AMECO and WEO databases; and Staff calculations.Note: The real-time and revised estimates are those produced in the Spring of years t and t+2 respectively.France - Revisions in GDP Potential Growth
(2005-2011)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: AMECO and WEO databases; and Staff calculations.Note: The real-time and revised estimates are those produced in the Spring of years t and t+2 respectively.3. This paper has a twofold aim. First, we update France’s potential GDP estimates provided by Cheng (2011). Second, looking forward, we relate the implementation of a more structural approach to fiscal policy to the robustness to new data of alternative de-trending methods.
4. The main lesson emerging from the analysis is that, while multivariate techniques improve robustness relative to univariate filters, they are also subject to revisions as new information becomes available. This means that, despite every effort to produce reliable, real-time potential growth estimates, uncertainty will inevitably remain a fact of life for policy makers. To make fiscal recommendations time proof, estimates for potential output should therefore take this uncertainty into account.
5. The rest of the paper is laid out as follows. Section II provides an overview of the methodologies covered, be they purely statistical or economic-based. Section III presents the estimates and projections of potential output for France during 1980-2018. Section IV calls for prudence in the use of potential output to inform fiscal advice. Section V concludes.
B. Cycle Extraction Methods
6. This study focuses on four de-trending methods. The Hodrick-Prescott and Christiano-Fitzgerald filters are purely statistical. The “production function” and “gaps” approaches combine filtering techniques with economic theory, therefore establishing a meaningful link between potential output and their underlying factors. This puts a premium on achieving consensus on the modeling and estimation methods to be employed. A brief description of each method is provided below.
The HP filter (Hodrick and Prescott, 1997): The trend GDP is the result of minimizing the distance between the trend and the original series as well as the curvature of the trend series. The exact trade-off between the two goals is judgmental. The filter suffers from asymmetry: with the exception of the central values, HP-filtered series are time-displaced compared to the underlying cycle, with revisions being particularly large for the most recent observations.
The CF filter (Christiano and Fitzgerald, 1999): This filter passes (rejects) frequencies within (outside) a certain range. It belongs to the class of band-pass filters, which can be either symmetric or asymmetric. Symmetric variants (e.g. Baxter-King filter) are free from phase shifts but this comes at the expense of series trimming. In contrast, CF uses the whole time series. CF outperforms the BK filter real time.
The PF approach (for EU countries see Denis et al., 2010): Instead of making statistical assumptions on the series properties of trends, PF makes assumptions based on economic theory, in particular the functional form of the production technology, trend technical progress (TFP), and the utilization of production factors. Below is a stylized representation of the PF methodology as currently applied by the EC (in collaboration with EU countries).
The gaps approach (Benes et al., 2010): To extract trend components, this method focuses on imbalances-type equations, specifically a Phillips curve, an Okun’s law, and a capacity utilization equation. Trend GDP, capacity utilization and the NAWRU are then made consistent through a multivariate filter. For the purposes of this paper, the overall estimate for potential GDP growth calibrated through the gaps approach is subsequently distributed across labor, capital, and total factor productivity via standard growth accounting techniques.
C. Results
7. Potential GDP growth in France declined markedly as a result of the crisis (Table 1; Figure 1). The downward trend is evident across methodologies. Potential GDP grew at an average rate of around 1.9 percent during the 1990s and 2000s, and decelerated to about 0.7 over 2008-12.
Potential GDP Growth, 1992-2012
Potential GDP Growth, 1992-2012
| Hodrick- Prescott | Christiano- Fitzegarld | Production Function, EC | Gaps Approach | Average across Methodologies | |
|---|---|---|---|---|---|
| 1992-2000 | 2.0 | 2.0 | 1.9 | 1.7 | 1.9 |
| 2001-2007 | 1.7 | 1.8 | 1.8 | 1.9 | 1.8 |
| 2008-2012 | 0.5 | 0.4 | 1.1 | 0.9 | 0.7 |
Potential GDP Growth, 1992-2012
| Hodrick- Prescott | Christiano- Fitzegarld | Production Function, EC | Gaps Approach | Average across Methodologies | |
|---|---|---|---|---|---|
| 1992-2000 | 2.0 | 2.0 | 1.9 | 1.7 | 1.9 |
| 2001-2007 | 1.7 | 1.8 | 1.8 | 1.9 | 1.8 |
| 2008-2012 | 0.5 | 0.4 | 1.1 | 0.9 | 0.7 |
France: Potential GDP Across Methodologies
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
France: Potential GDP Across Methodologies
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
France: Potential GDP Across Methodologies
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
8. Economic approaches can identify the channels through which the crisis has influenced potential output. By the gaps approach, around ¾ of the shortfall in potential growth during 2008-12 relative to 1992-2007 would come from a declining TFP, with labor and capital contributing about evenly to the remaining shortfall (Figure 2).
France: Factors Underlying Potential GDP Growth, Gaps Approach
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: Staff calculations.France: Factors Underlying Potential GDP Growth, Gaps Approach
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: Staff calculations.France: Factors Underlying Potential GDP Growth, Gaps Approach
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: Staff calculations.TFP: The crisis would have aggravated the capacity of France firms to innovate and remain competitive both in domestic and abroad. The production switch initiated in the past from the relatively high-TFP growth manufacturing sector towards services would also weigh down on overall productivity (see, e.g., Molagoda and Perez-Ruiz, 2011).
Labor: A somewhat less dynamic labor force, together with a higher rate of structural unemployment, would have pushed the labor contribution downwards. Indeed, Bonthius et al. (2013) provide recent evidence of an outward shift in France’s Beveridge curve and a higher structural unemployment.
Capital: Lower capital accumulation would reflect heightened risk aversion amongst entrepreneurs, weak demand prospects, as well as more cautious bank lending.
9. The estimated gaps for GDP, the unemployment rate, and capacity utilization are closely linked together. As predicted by Okun’s law, the unemployment gap is strongly correlated with the current and lagged output gap. It is also apparent that the unemployment gap has smaller cyclical fluctuations. This would reflect well-known features of the labor market, such as labor hoarding and the discouraged-worker effect during recessions. The capacity utilization gap displays pronounced volatility, the decline during recessions being particularly steep.
Gaps Approach - Estimated Gaps
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: Staff calculations.Gaps Approach - Estimated Gaps
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: Staff calculations.Gaps Approach - Estimated Gaps
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: Staff calculations.10. The loss of potential GDP levels induced by the crisis thus far could be substantial. The shortfall in potential GDP levels could have reached around 4 percent by 2011 (10 percent cumulatively over 2008-11). These losses have significant implications in terms of living standards, and create strains for public finances. In this connection, it is important that supply-side policies aim at minimizing such losses going forward. Section V illustrates the scope for policy reform. Before we turn to that, we discuss prudence as a main operating principle to inform fiscal policy recommendations in Section IV.
D. Potential GDP Estimates: Dealing With Uncertainty
11. We take potential GDP by the gaps approach as our baseline scenario. Following a marked, transitory downturn in potential growth rates, and a gradual medium term recovery, it is expected that potential growth increases from about 0.9 during 2012-13 to 1.4 percent in 2018 (the last WEO projection year), and to 1.6 (average growth rate 1991-2012) by 2021.
12. To understand the gains from using the gaps approach to estimate and project potential GDP, it is helpful to recall the two major problems with univariate filters: they ignore relevant economic information, which can create biases; and they give excessive weight to revisions to the most recent observations. To make this clear, Section A discusses the performance of both univairate and multivariate techniques in terms of robustness to new data. Still, it is important to stress that, just as any other method to estimate potential GDP, the gaps approach also rests on a number of choices (Section B and Section C).
Robustness to New Information
13. The potential GDP estimates presented in this study, based on multivariate filtering techniques, are found to be more stable than alternative univariate methods. To illustrate this point, we examine the performance of HP, CP, and the gaps approach as news arrives to the set of information used for projections. This comprises new data for GDP (HP and CF filters), along with data on capacity utilization and the unemployment rate (gaps approach). The performance evaluation excludes the EC’s production function methodology because the exact specification of the NAWRU and TFP extraction methods, two key inputs for calculating potential GDP, are unavailable to staff.
14. We inspect the prediction error over one-to-six quarters ahead. The revisions for each quarter t are therefore defined as:
where Ri,t is the i-th revision of observation t, and ypt,t+i is the estimate of potential output for period t, by using information up to period t+i. We then obtain the root mean square revision (RMSR), a standard measure of predictive power, by aggregating the forecast errors up to six quarters ahead across all observations, that is
RMSR is therefore the square root of the mean of the squares of the prediction errors Ri,t for horizons 1 through 6 quarters ahead.
15. We report RMSR for both the level and the growth rate of potential GDP (table 2). As expected, all three methods have increasing revisions over time. The revisions grow steeply between the first and second quarter. This may reflect the timing of national accounts releases: each quarter’s estimates can be significantly improved upon the release of the flash estimates for such quarter (realized at t+45).
Root Mean Square Revisions
Root Mean Square Revisions
| Out-of-sample revisions at quarter… | ||||||
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | |
| Level of potential GDP | ||||||
| Gaps Approach | 0.2 | 0.3 | 0.4 | 0.5 | 0.6 | 0.6 |
| Hodrick-Prescott | 0.2 | 0.5 | 0.6 | 0.8 | 0.9 | 1.0 |
| Christiano-Fitzgerald | 0.3 | 0.5 | 0.6 | 0.8 | 0.9 | 0.9 |
| Annualized growth rate in potential GDP | ||||||
| Gaps Approach | 0.1 | 0.1 | 0.1 | 0.1 | 0.2 | 0.2 |
| Hodrick-Prescott | 0.1 | 0.2 | 0.2 | 0.3 | 0.4 | 0.4 |
| Christiano-Fitzgerald | 0.1 | 0.2 | 0.2 | 0.3 | 0.3 | 0.4 |
Root Mean Square Revisions
| Out-of-sample revisions at quarter… | ||||||
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | |
| Level of potential GDP | ||||||
| Gaps Approach | 0.2 | 0.3 | 0.4 | 0.5 | 0.6 | 0.6 |
| Hodrick-Prescott | 0.2 | 0.5 | 0.6 | 0.8 | 0.9 | 1.0 |
| Christiano-Fitzgerald | 0.3 | 0.5 | 0.6 | 0.8 | 0.9 | 0.9 |
| Annualized growth rate in potential GDP | ||||||
| Gaps Approach | 0.1 | 0.1 | 0.1 | 0.1 | 0.2 | 0.2 |
| Hodrick-Prescott | 0.1 | 0.2 | 0.2 | 0.3 | 0.4 | 0.4 |
| Christiano-Fitzgerald | 0.1 | 0.2 | 0.2 | 0.3 | 0.3 | 0.4 |
16. The gaps approach outperforms both HP and CF. As well documented in the literature, the HP filter has a tendency to attribute part of the cycle to the trend, given its excessive sensitivity to the last observations in the sample. Multivariate techniques such as the gaps approach are partly immune to the end-point bias problem as they can extract the information about the cycle from additional, observable indicators strongly linked to output (capacity utilization and unemployment in Benes et al., 2010).
17. Recent methodological improvements to the PF approach adopted by the EC have sought to reduce the size of forecast errors. For instance, the size of the revisions at one-year horizon from switching to a Kalman filter TFP extraction were reduced as much as 20 percent for most countries (Denis et al., 2010). Older and Kalman-filter-based estimates differ the most at turning points—the periods where precision is most crucial for policy makers.
Choices
18. The set of choices embedded in the estimates presented in this paper fall under three main categories: (i) assumptions affecting the exact specification of the Phillips curve, the Okun’s law, and the capacity utilization equation; (ii) assumptions affecting the smoothness properties of potential GDP; and (iii) assumptions on the account of the impact of policy reforms (in train or announced). This section discusses (ii) and (iii). For a detailed description of the model equations the reader is referred to Benes et al. (2010).
Smoothness properties of potential GDP. The same approach can deliver higher or lower variability of potential output depending on the relevant time horizon under consideration. The longer the reference time horizon is, the less production factors are affected by cyclical fluctuations relative to structural drivers. To illustrate, Central banks, whose primary purpose is the prevention of inflationary tensions, will tend to favor shorter horizons and relatively flexible potential outputs. By contrast, if the main focus is to target structural deficit targets, we would give preference to a longer time horizon and a low-volatility trend. To serve this latter purpose, we choose a cut-off periodicity of 56 quarters to separate the short- and long-term fluctuations: the trend is defined in such a way that more long-term movements are captured in it.
Treatment given to policy reforms. An important issue when estimating potential GDP is whether policy effort is incorporated to the underlying trend series, particularly the NAWRU. Overall, staff generally views the recently adopted reforms (the reduction in the tax wedge pursued by the Crédit Impôt Compétitivité Emploi (CICE); the move towards “flexicurity” embedded in the labor market reform; and the expansion of employment subsidies)2 as a good basis for lifting France potential GDP. These reforms are, however, not fully incorporated in the estimates presented in this paper as uncertainties remain about their timing and implementation—insofar as their success critically hinges upon behavioral changes by social partners, enterprises, or even the judiciary.
In estimating potential output, an important consideration should be the impact of estimation errors on policy. If the cost of projection errors is asymmetric (i.e., high in the event of an overvaluation of potential output), this would argue for using more conservative assumptions. This consideration is relevant, for instance, when potential growth estimates are used to anchor fiscal policy, since the cost of overestimating potential output is a loss of fiscal space which is harder to adjust to than the opening of additional fiscal space (in the event of an underestimation of potential output). Having estimated the cumulative fiscal adjustment required under prudent assumption, there is then room to decide on a speed and profile of adjustment that takes into account cyclical conditions.
Permanent or Temporary Effects from the Crisis? Probably Temporary
19. Conceptually, the damage of a crisis on an economy’s potential can be either temporary or permanent. Do recessions shrink the pace of growth only temporarily or more durably? The length of the downturn is critical in this respect: a protracted recession can be expected to affect the pace of growth persistently through hysteresis effects. Focusing on labor, for instance, the longer people are out of work, the more likely their skills will fade or become less suitable for the available jobs. Thus, high levels of long-term unemployment may portend high levels of structural unemployment.
20. It is challenging, however, to distinguish in real time permanent versus temporary effects. The 2013 WEO proposes to use inflation as an instrument to separate temporary vs. permanent effects. Inflation stickiness, the argument goes, could be indirect evidence that output gaps are small, the observed increase in unemployment mostly structural3, and potential GDP much lower than prior to the crisis. The competing view is that most of the economy’s potential has been preserved throughout the crisis, and, consequently, the output gap is currently sizable. This can, in turn, be reconciled with a relatively stable inflation provided that the Phillips curve has flattened (inadequate responsiveness of inflation to economic fluctuations in a low inflation environment) and/or that inflation expectations remain well-anchored.
21. The potential GDP estimates presented in this paper are consistent with the view that France will largely recoup crisis-induced losses in potential GDP growth (i.e., the losses in growth are temporary but the loss in the level of output is permanent). The inflation projections embedded in WEO are set to decline marginally this and next year and gradually converge to the 2 percent anchor thereafter. This presumes a low sensitivity of wages to unemployment conditions (i.e. a relatively flat Phillips curve).
France - Output Gap and CPI-Inflation
(Annual Percentage Change)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: European Commission (EC), France Stability Program (SP), WEO, and StaffFrance - Output Gap and CPI-Inflation
(Annual Percentage Change)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: European Commission (EC), France Stability Program (SP), WEO, and StaffFrance - Output Gap and CPI-Inflation
(Annual Percentage Change)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: European Commission (EC), France Stability Program (SP), WEO, and StaffCalling for Action: Medium-Term Prospects and Policy Scenarios
22. This section gauges the impact on potential growth of alternative paths for its underlying drivers. These simulations are meant to illustrate the scope for higher potential growth from structural reforms that would operate through these channels. The assumed shocks would get France closer to best performers on key growth drivers in the way described below.
TFP shock. France productivity growth pre-crisis (av. 2000-08) stood at 0.7 percent per year, lagging behind innovation leaders, including Sweden and Finland (1.7 percent each); and the UK (1.4 percent). All three countries outperform France by the World Bank Doing Business and the Global Competitiveness indicators. Reducing the administrative burdens and tax complexity, and enhancing competition in services are some of the levers to influence TFP (see companion Selected Issues Paper on “France: External Developments and Assessment”). The simulated scenario would allow France to close by 2018 half of the gap between the current level of TFP growth and the 2000-08 average rate across best performers (Sweden, Finland, and the UK). This amounts to a yearly positive TFP shock of 0.1 percent point relative to the baseline.
Participation shock. Workforce participation in France stood at around 71 percent of the working age population in 2011–4 percentage points below the OECD average. The gap with best performers is at its highest for the 55-64 years old cohort (51.2 percent against around an average of 73 percent for Germany, Denmark, Finland, Sweden, and Switzerland). The implicit tax on continuing work (in terms of contributions paid and foregone benefits) is still high in France relative to peers. As in the “high participation” scenario by INSEE (Filatriau, 2011), we simulate a shock where each year the labor force is, on average over 2013-18, 0.65 percent above the baseline. This portrays a situation with enhanced participation from both older and young workers.4 Instruments to increase employment rates include the forthcoming pension reform, and the reform of unemployment insurance. The duration of unemployment spells could also be lowered by reducing mobility costs both across regions (cost of services and housing) and professions (consolidation of multiple pension and benefit systems).
Structural unemployment shock. On the eve of the crisis, France’s structural unemployment reached almost 9 percent of the labor force, largely exceeding the NAWRU featured by labor markets with greatest “flexicurity”, such as those of Denmark (5.2), Netherlands (3.6), Sweden (6.3), or Finland (6.9). Continued efforts to reduce tax wedges along with employment-friendly wage policies (temporary freeze of the minimum wages), lower dismissal costs) are potential instruments to curb structural unemployment. The simulated scenario assumes a yearly ¼ percent point reduction in the NAWRU relative to the baseline over 2013-18. This would allow France to cover half of the distance with best performers.
Labor Force- Baseline and Policy Scenario
(In thousands)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: INSEE, 2011.1/ Young and old workers are, respectively, 25 years old or less and 55-59 years old.Labor Force- Baseline and Policy Scenario
(In thousands)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: INSEE, 2011.1/ Young and old workers are, respectively, 25 years old or less and 55-59 years old.Labor Force- Baseline and Policy Scenario
(In thousands)
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Sources: INSEE, 2011.1/ Young and old workers are, respectively, 25 years old or less and 55-59 years old.23. The simulations are indicative of substantial gains from reform (Figure 3). By 2018, potential growth would outpace that embedded in the baseline by between around 0.1 and 0.4 percentage points, depending on the scenario under consideration. The combined impact of those changes would raise potential growth to around 1.8 percent by 2018 (against 1.4 percent in the baseline).
France: Simulations on Potential GDP
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: Staff calculations.France: Simulations on Potential GDP
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: Staff calculations.France: Simulations on Potential GDP
Citation: IMF Staff Country Reports 2013, 252; 10.5089/9781484389133.002.A002
Source: Staff calculations.Conclusion
24. This paper has updated projections for France potential GDP, discussed its main drivers, and the scope for policy reform. We have also evaluated the real-time performance of alternative methods to estimate potential output—the HP and CF filters, and the termed “gaps methodology”.
25. The main lesson is that potential growth estimates are inevitably subject to revisions as new data become available. Therefore, uncertainty about the true level of potential output will remain a fact of life for policy makers. There will always be a need to periodically fine-tune the method with developments in the literature, new data sources, and alternative estimation approaches. Under these conditions, it is important that the methodology underlying potential growth estimates (and its limitations) be well understood. Given this inherent uncertainty, it is particularly important to rely on assumptions, parametric choices, and estimation techniques that do not expose policies to costly mistakes. For fiscal policy, this would argue in favor of conservative assumptions.
26. The structural approach to potential growth projection also allows to gauge the potential impact of structural reforms which can increase the contributions of the various factors of production. France has already embarked on structural reforms that could trigger some of these increases, provided these reforms are sustained and deepened. This paper illustrates how a broad-based structural reform program aimed at closing the gap relative to best performers (in terms of labor force participation, structural unemployment, and total factor productivity) could increase potential growth by as much as 0.5 percent on average over 2013-18.
References
Benes, J., K. Clinton, R. Garcia-Saltos, M. Johnson, D. Laxton, P. Manchev and T. Matheson, 2010, “Estimating Potential Output with a Multivariate Filter,” IMF Working Paper 10/285.
Bonthuis, B., V. Jarvis, and J. Vanhala, 2012, “What’s going on behind the euro area Beveridge Curve,” manuscript.
Cheng, K., 2011, “France’s Potential Output during the Crisis and Recovery,” IMF, France, Selected Issues Paper, IMF Country Report No. 11/212.
Denis, C., K. Havik, K. Mc Morrow, C. Planas, R. Raciborski, W. Röger, and A. Rossi, 2010, “The production function methodology for calculating potential growth rates and output gaps,” European Economy, Economic papers, No. 420, European Commission, Brussels.
Filatriau, O., 2011, « Projections à l’horizon 2060: Des actifs plus nombreux et plus âgés », INSEE Première n° 1345, avril 2011.
Lawrence J. C., and T. J. Fitzgerald, 1999, “The Band Pass Filter,” NBER Working Papers 7257, National Bureau of Economic Research, Inc.
Hodrick, R., and E. C. Prescott, 1997, “Postwar U.S. Business Cycles: An Empirical Investigation,” Journal of Money, Credit, and Banking, 29 (1), 1–16.
WEO, 2013, “Hopes, realities, and risks”, International Monetary Fund, Washington D.C.
Prepared by Jaromir Benes and Esther Pérez Ruiz.
See Staff Report for the 2013 Article IV consultation with France for a detailed description of these reforms.
If those who are unemployed cannot effectively compete for jobs, they may have much less influence on the wages of those who are employed. This can translate into less influence on the prices firms charge for their goods and services. Such unemployment is termed “structural”.
The 2010 pension reform is already incorporated in the INSEE central scenario used in our central projections.
