Диссертация (1137741), страница 14
Текст из файла (страница 14)
However,when the share of borrowers exceeds the threshold level the multiplier becomes negative.64Chapter 3. Optimal fiscal policy under an unbalanced pensionsystem and the role of non-Ricardian agents503.1.IntroductionA wide range of pension reforms has been launched in many countries in order toensure the sustainability of the social security system, which is at risk because of increasingexpenditure on pensions.51 The main drivers of these dynamics are demographic changes:fertility rates which are below the replacement level and increasing life expectancy. Thesedemographic developments are leading to a significant increase in the old-age dependencyratio: the ratio is expected to rise from 29.9 in 2016 to 52.2 in 2070 in the EU-27, while thevery old-age dependency ratio will increase from 8.4 to 23.0 during this period, according torecent forecasts from the European Commission.
52 Despite the implemented reforms, federaltransfers remain one of the key sources to balance the budget for pension funds. Thesetransfers are expected to rise steadily in the future: this part of fiscal expenditure will increasein OECD countries from 8.9% of GDP in 2013-2015 to 9.5% of GDP in 2050. 53 In Russia,public expenditure on the pension provision is projected to increase from 6.7% of GDP in2017 to 7.5% of GDP by 2035 according to the estimates provided in Chapter 1. 54Pension reforms can be considered as an alternative to the traditional measures of fiscalconsolidation. This study defines the optimal fiscal instruments (rate of social contributions,income tax, government expenditure) chosen by the social planner under an unbalancedpension system in the economy with Ricardian and non-Ricardian agents (rule-of-thumbconsumers),55 who consume all their disposable income each period.
First, we specify howoptimal social contributions and income tax change with the retirement age, life expectancy,and labor productivity, and how they depend on the type of pension system (balanced orunbalanced) in the absence of non-Ricardian agents. Second, we define the choice of socialcontributions, income tax and government expenditure by a benevolent government in an50Some of the results in the Chapter were presented in Mamedli, M. O. (2017).
Fiscal Policy and the UnbalancedPension System, HSE Economic Journal, vol. 21, no 1, pp. 114–144.51The overview of the reforms is provided in Chapter 1.52European Commission, The 2018 Ageing Report: Underlying Assumptions and Projection Methodologies,Nov. 2017. Very old-age dependency ratio refers to 80+/15-64.53OECD, Pensions at glance 2017.54The estimates are provided for the case without an increase in retirement age, economic growth equal to 1.5%and with pension indexation for working retirees.
Chapter 1 provides the forecasts of public spending onpensions under different economic and demographic assumptions.55See Galí et al. (2007).65economy with both Ricardian and non-Ricardian agents. The focus is on how optimal policyinstruments change with the share of non-Ricardian consumers.Models with non-Ricardian agents have become common in the analysis of fiscalpolicy.
The inclusion of population heterogeneity via Ricardian and non-Ricardian consumersallows us to capture the failure of Ricardian equivalence. 56 There are several papers thatanalyze structural fiscal reforms and demographic shocks within the Global IntegratedMonetary and Fiscal model (GIMF)57, which incorporates both Ricardian and non-Ricardianconsumers (Almeida et al.
2013a, b; Castro et al., 2015;58 Karam et al., 201059). However, asopposed to the current research they do not consider the optimal choice of instruments by apublic planner and the impact of the share of non-Ricardian consumers in this choice.The research on pension reforms can be classified by the type of pension system underconsideration. The first group consists of research on PAYG reforms (e.g. Nickel et al., 2008;Karam et al., 2010; Kilponen et al., 2006 60; Castro et.
al., 2017; Almeida et al., 2013 a, b;Pierrard, Snessens, 2009; Marchiori, Pierrard, 2012, 201561). This research falls into the firstcategory. Other research considers the switch from pay-as-you-go (PAYG) to a fully fundedpension system (Borsch-Supan et al., 2006; McGrattan, Prescott, 2017). Both types of pensionsystems were analysed in Marchiori et al. (2011) and de la Croix et al. (2013).
There are alsoseveral studies (Galasso, 2008; 62 Heijdra, Romp, 200963; Beetsma et al., 2013 64) whichconsider an optimal pension plan or optimal retirement age.56The impact of fiscal policy shocks on private consumption can be captured by the introduction ofheterogeneity in consumers (Galí et al., 2007).57The baseline structure of model is based on GIMF, presented in Kumhof, Laxton (2007, 2009, 2013), Kumhofet al.
(2009, 2010).58Almeida et al. (2013a) extend the Portuguese Economy Structural Small Open Analytical model incorporatingfinancial frictions (as in Bernanke et al., 1999). The model is applied in research on the assessment of structuralreforms, fiscal stimulus and consolidation (Almeida, Castro, Felix, 2009, 2010; Almeida et al., 2010a, b).59Karam et al. (2010) consider three reforms of the PAYG pension system: an increase in the retirement age,cuts in pensions and an increase in social contributions.
They show that pension reforms can have a positiveeffect both in the short run, due to an increase in consumption, and in the long run, due to the lower level ofpublic debt.60Kilponen et al. (2006) extend the OLG model of Gertler (1999) introducing distortionary taxation, timevarying retirement and death probabilities to the model. The model is used for the analysis of demographicchanges in a small open economy, using the example of Finland.61Pierrard and Snessens (2009) and a series of further research studies (Marchiori and Pierrard 2012; 2015)sequentially extend the OLG model of Auerbach and Kotlikoff (1987). The model is calibrated for Luxembourg(Luxembourg Overlapping generation model for policy analysis).62Galasso (2008) within a calibrated model of Galasso and Profeta (2004) extended it to incorporate anendogenous retirement age by conducting simulations up to 2050 for France, Italy, the United Kingdom and theUSA, showing that the retirement age and social contributions rate would be increased in all countriesconsidered except Italy.63Heijdra and Romp (2009) analyzed the households’ retirement age decisions in the model with exogenousinterest rate and endogenous labor supply (developed by Heijdra and Romp, 2006; 2008).
They show that forpension reform to have an impact on the households’ decisions, the change needs to be significant.66The analysis in this chapter is based on the overlapping generations model (OLG) ofHeijdra and Bettendorf (2006) who extended the framework initially developed by Yaari(1965) and Blanchard (1985). In order to investigate the optimal policy mix of socialcontributions and income tax rate, we extend the model of Heijdra and Bettendorf (2006).They analyzed the economic consequences of lower pensions and a higher retirement age inan open economy with traded and non-traded sectors (as in Turnovsky, 1997). 65 However,Heijdra and Bettendorf (2006) consider an exogenous interest rate along with a rudimentarybalanced pay-as-you-go (PAYG) pension system.
66 While Heijdra and Bettendorf (2006)consider the consequences of shocks to the welfare of each generation, we investigate theoptimal set of measures conducted by a benevolent government, which maximizes the socialwelfare function, and how this set of measures is affected by the share of non-Ricardianconsumers. In this chapter, their model is extended to investigate the unbalanced budget ofthe pension fund in a closed economy with an endogenous interest rate.
This allows us toaccount for the effect of different economic characteristics (retirement age, life expectancy,labor productivity) on capital accumulation. The model is extended to incorporate bothRicardian and non-Ricardian consumers in order to investigate how the share of nonRicardian consumers affects the choice of policy instruments. Following Heijdra andBettendorf (2006) the model also incorporates decreasing-with-age labor productivity whichadds non-Ricardian characteristics to the model.Heijdra and Bettendorf (2006) consider two types of demographic shocks: aproportional decrease of both birth and death rates so that population growth remainsconstant, and a fall of the birth rate accompanied by a growth of the death rate, which leads toa decreasing population size.
They consider two types of pension reforms: a decrease inpensions and an increase in the retirement age. Both reforms are followed by an adjustment insocial contributions because Heijdra and Bettendorf (2006) consider a balanced pensionsystem. If the interest rate exceeds population growth, both reforms lead to an increase inconsumption and financial assets in the long-run, yet also to a decrease in capital, becausethere is a higher consumption of non-tradable goods.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
