Output gap, participation and minimum income: a proposal for Italy

Abstract

Recently, some attempts to increase the stance for fiscal policies in the European budgetary framework have followed the line of reducing the estimated “structural” unemployment rate (NAIRU), with the ensuing increase in the computation of the output gap. A similar effect can be obtained by increasing the actual participation rate. In this paper, we propose the introduction of a deficit-financed conditional minimum income (CMI) to discouraged people which are outside the labor force. By stimulating participation, this measure would bring about an upward revision of Italy’s potential output, and this in turn will contribute to generate a greater fiscal stance. We empirically assess the reliability of this measure by using both comparative statics and empirical estimations carried out via the simulation procedure used by the Output Gaps Working Group of the European Commission. Assuming one million newcomers in the labor force, our findings indicate that the measure would have produced a greater fiscal space of approximately €19 billion in 2016 and €12 billion in 2017. We also forecast the impact of the introduction of the deficit-financed CMI on real GDP and public finance indicators. We finally discuss the feasibility and the main criticisms of the proposal.

Keywords:

• Aggregate demand
• fiscal stance
• minimum income
• NAIRU
• potential output

JEL CLASSIFICATION:

• E27
• E62
• J08

Acknowledgment

This research was conducted as part of the works of the Jean Monnet Center of Excellence on “Labour, Welfare and Social Rights in Europe” hosted by Roma Tre, Department of Economics. A preliminary draft of this paper has been presented at the 30th EAEPE Conference (Nice, 6–8 September 2018). We thank all the participants for useful comments. Moreover, we are grateful to Sebastiano Fadda, Riccardo Pariboni, Luigi Salvati and Antonella Stirati, for fruitful discussion. Last but not least, we thank the editor and two anonymous referees for valuable remarks and suggestions. The usual disclaimer applies.

Notes

1 On January 2020, Eurostat indicated a positive but moderate improvement concerning Euro area (EA19) unemployment rate (7.4%). Importantly, the reader should take into account that the present work has been conceived in 2018 and developed before the outbreak of the Covid-19 emergency.

2 See on this Jarociński and Lenza (2018).

3 It is worth pointing out that in the EU fiscal framework reference is made to the NAWRU (non-accelerating wage rate of unemployment) instead of the NAIRU when assessing the ‘natural’ or ‘structural’ rate of unemployment. While the latter refers to the traditional price inflation features in the Phillips curve, the former refers to wage inflation. Despite this difference, they can be interpreted in the same way: when the actual unemployment rate is below this ‘equilibrium’ rate, inflationary pressures would take place. Throughout the article, we shall use the terms NAIRU and NAWRU interchangeably.

4 See for example the note by Adam Tooze published on 30th April 2019 by Social Europe (website: www.socialeurope.eu/output-gap-nonsense). The reader may also refer to the works by Robin Brooks and Greg Basile published by the Institute of International Finance (website:www.iif.com).

5 The interview was released at Le Figaro. See AGI (2016).

6 In addition to Padoan, also Cacciotti, Frale, and Teobaldo (2013), Cottarelli, Giammusso, and Porello(2014), Fioramanti (2016), Frale and De Nardis (2017), and MEF (2017a) posed similar questions to the output gap estimations for Italy.

7 Following Keynesian insights, fiscal policy is supposed to be countercyclical, hence a greater fiscal stance is allowed when the economy stands well below its potential.

8 Putting it simply, the methods of computation of potential GDP can be clustered in two families: those ones based on the use of filtering techniques (implying the smoothing of actual GDP data), and those ones based on the production function approach (grounded on the estimation of potential levels of the factors of production). For a broader overview of these methods, see Cerra and Saxena (2000) and CBO (2004). Recently, some challenging attempts to introduce innovative techniques—or interesting modifications to the existing ones—for the computation of potential GDP have been proposed by Fontanari, Palumbo, and Salvatori (2020), Charles, Dallery and Marie (2018) and Li and Mendieta-Muñoz (2018).

9 Equation (2) is homogeneous of degree 1. Since constant returns to scale and perfect competition are assumed, this functional form does not allow for the presence of any ‘residual’ in the distribution of product. Moreover, being in this framework marginal products equal to their remuneration (in physical terms), the parameter $\alpha$ should reflect the labor share on GDP. In the EC model for the computation of potential GDP, it is fixed for all EU countries at 0.65 (see Fioramanti 2016).

10 The semi-elasticity captures the reactivity of the budget balance (in % of GDP) to a change in the output gap. Its value stems from a weighted average of the individual elasticities of each government revenue and spending item. For technical issues, see Mourre, Astarita, and Princen (2014) and Price, Dang, and Guillemette (2014). Recently, Mourre, Poissonnier, and Lausegger (2019) have offered an update and further analyses on the estimation of this parameter.

11 The EC yearly analyses in detail EU member states’ economic and structural reforms programmes and provides them with a set of recommendations for the subsequent 12–18 months. During this time, each member state has to align their budgetary and economic policies with the objectives and rules agreed at EU level, within the previously mentioned targets. For a more detailed picture, see EC (2017).

12 For a comprehensive overview of the Italian context, see Cesaratto and Zezza (2018) and Storm (2019).

13 After the outbreak of the pandemic emergency, the number of beneficiaries dramatically increased up to 3.1 million people, that is 1.3 million households (September 2020). Source: INPS.

14 See MEF (2019, p. 33) for further details.

15 For a critique on the concept of NAIRU, see also Stirati (2001), Sawyer (2002) and Storm and Nastepad (2007); Girardi, Paternesi Meloni, and Stirati (2020) provide some empirical support against the independency of potential GDP of aggregate demand.

16 A certain degree of pro-cyclicality has been explained through the presence of labor market rigidities (Rusticelli 2015). In parallel, the ‘hysteresis hypothesis’ has been advanced to describe labor market ‘anomalies’ in European Countries (Blanchard and Summers 1986; Havik et al. 2014).

17 To this regard, Girardi, Paternesi Meloni and Stirati (2020) have recently estimated the effects of demand impulses on key macroeconomic outcomes, finding that autonomous demand shocks fosters GDP, lowers unemployment and upsurges other factors which may stimulate potential growth (such as productivity, participation, and capital stock).

18 As in the scheme of the RDC, the sum represents the minimum monthly income needed to reach, on average among regions, the reference ‘absolute poverty’ threshold in Italy (cf. ISTAT 2018).

19 For the third quarter of 2018, the official statistics on Italy indicate a vacancy rate of 1.1% in industry and services sector, which is approximately equal to 270 thousand vacancies.

20 Our policy proposal would not imply a reduction in unemployment benefits since people involved would be outside the labor market, and hence they are not be getting any unemployment benefits.

21 One can argue that a low-paid worker can be encouraged to quit the job in case the CMI is higher than the salary. While the monitoring of the precedent job can formally mitigate this issue, the introduction of a minimum wage level at least equal to the CMI would discourage these behaviors.

22 Source: ISTAT, potential labor force. Even if the most common data by ISTAT concern the age group 15–64, we used the 15–74 statistics to be consistent with EC procedures.

23 In case the scope of measure would not be enough to involve all discouraged individuals, a selection system would be necessary to identify the beneficiaries. For instance, this mechanism may take into consideration the economic condition of the candidates, by given priority to the poorest ones.

24 The simulation, which results are reported in Table 1, starts from 27,317.6 million workers in the total labor force.

25 To be fair, also the actual unemployment rate would increase, but this would not significantly impact the real conditions of involved people since they would pass from inactive to unemployed, with no changes on their personal situation. The expansionary fiscal policy is however supposed to stimulate the economy and then to foster employment opportunities, as we will show in the simulation proposed for 2017.

26 Estimates of potential GDP have been computed via the RATS routines made available by the OGWG at the CIRCABC public repository. We have followed the European Commission’s methodology which entails the estimation of TFP—in its trend and cycle components—and NAWRU through the GAP package, that is also publicly available at the CIRCABC repository.

27 Basically, the procedure we follow in the simulation allow us to consider as ‘endogenous’ all the variables in our empirics, then taking into account to feasible side-effects of the increasing participation rate on TFP and the NAWRU. See fifth section (second subsection) for a discussion.

28 Discrepancies in participation rates with the previous exercise would depend on the fact that the simulation for 2017 considers only the civilian labor force.

29 According to the existing fiscal framework, Italy still has to cope with a prospective structural deficit of 0.5% GDP, that is the medium-term budgetary objective.

30 Being our stimulus a combination of subsides for the beneficiaries and public effort in reinforcing the job centers, we refer to the fiscal multiplier of total expenditure calculated by MEF (2017b).

31 In the simulation reported in Table 2, for 2017 we start from 25.7 million workers involved in the civilian labor force. The simulation has been carried out by adding one million persons to the labor force and the unemployment. This increases the unemployment rate up to 14.62%, whose value settles at 14.32% if we consider the expansionary effect of the measure. See Table 3 for details.

32 See the interview to Daniel Gros reported by Paudice (2018). The interview refers to the work by Tridico and Paternesi Meloni (2018), where the general features of this proposal have been introduced.

33 In each year t, the main economic and fiscal data presented in the Stability and Convergence Programmes entail the previous year (year t-1), the current year (year t) and at least the following three years (year t+1 to t+3). The compliance of each member state is assessed on an ex-post basis for the year t-1, an in-year basis for the year t and on an ex-ante basis for years from t+1 to t+3.

34 As a matter of fact, the NAIRU theory has become the prevalent explanation of unemployment in the European framework, and it is often used to advocate policies aiming at reducing welfare provision (particularly unemployment benefits) to decrease actual unemployment rates (cf. Stockhammer 2008). However, this view has been criticized by authors who endorse a different perspective on the inflation-unemployment nexus (see Sawyer 2002; Arestis and Sawyer 2005; Stirati and Paternesi Meloni 2018). According to this alternative approach, output and employment essentially depend on aggregate demand also in the long run, while income distribution can be affected by the bargaining power of parties. See Stirati (2001) and Serrano (2006) for an interpretation of inflation and hysteresis according to these lines of interpretation, as well as Kurz and Salvadori (1995) and Petri (2004) for advanced expositions of the analytical background.

35 See German Council of Economic Experts (2015/16), Focus on Future Viability, Appendix, p. 34.

Additional information

Notes on contributors

Giacomo Bracci

Giacomo Bracci is a PhD student at Dipartimento di Economia e Management, Università di Trento, via Virgilio Inama, 5, 38122 Trento.

Walter Paternesi Meloni

Walter Paternesi Meloni is a Post-doctoral Research Fellow at Department of Economics, Roma Tre University, via Silvio D’Amico, 77, 00145 Rome.

Pasquale Tridico

Pasquale Tridico is a Full Professor at Department of Economics, Roma Tre University, via Silvio D’Amico, 77, 00145 Rome.