An overlapping generations general equilibrium model of Croatian pension system

Cilj doktorskog istraživanja je odrediti putem kojih strukturnih i parametarskih reformi se postiže optimalan način financiranja hrvatskog mirovinskog sustava. U radu se analiziraju i makroekonomski učinci i utjecaj na blagostanje stanovništva, uzimajući u obzir suprotstavljene ciljeve održivosti mirovinskog sustava i adekvatnosti mirovina. Znanstveni doprinos proizlazi iz uniformne evaluacije ekonomskih i socijalnih učinaka reformi unutar modela opće ravnoteže preklapajućih generacija, kako bi se zaključilo koja od pet predloženih alternativa se sveukupno može ocijeniti najboljom za sustav. Primijenjena metodologija nadograđuje Diamondov (1965) neoklasični OLG model, omogućavajući proširenje na tri generacije, državu i dvostupni mirovinski sustav. Nakon uspostave teorijskog okvira OLG modela i analize specifičnosti hrvatskog mirovinskog sustava, provedena je empirijska implementacija OLG modela hrvatskog mirovinskog sustava uz pomoć metode kalibracije. Empirijskim istraživanjem utvrđeno je da jačanje individualne kapitalizirane mirovinske štednje ima negativan učinak na fiskalnu poziciju gospodarstva, čime je potvrđena prva hipoteza, ali pozitivno utječe na akumulaciju kapitala i agregatne štednje, gospodarski rast te blagostanje stanovništva. Suprotno, jačanjem sustava generacijske solidarnosti postiže se povećanje održivosti mirovinskog sustava, no njegov negativan učinak na veći broj makroekonomskih kategorija umanjuje pozitivne implikacije, zbog čega je djelomično prihvaćena druga hipoteza. Nadalje, reformom smanjenja ukupnih mirovinskih doprinosa ostvaruje se rast osobne potrošnje i usporavanje pada bruto privatnih investicija, čime se djelomično potvrđuje treća hipoteza. Za razliku od toga, povećavanje izdvajanja za prvi mirovinski stup poboljšava fiskalnu poziciju gospodarstva, ali u nedovoljnoj mjeri da se zatvori jaz između mirovinskih izdataka i doprinosa, a ostvaruju se negativni učinci na blagostanje pa se potvrđuje četvrta hipoteza. U uvjetima starenja stanovništva, DC sustav je superiorniji od DB sustava, ali su obje alternative inferiorne u usporedbi s trenutnim sustavom temeljenom na mirovinskim bodovima. Istraživanjem je utvrđeno da buduće reforme trebaju ići u pravcu parametarskih reformi prvog mirovinskog stupa i smanjenja doprinosa za mirovinsko osiguranje.Reforms of the Croatian pension system have always been politically controversial and subject to intense public debate. The existing pay-as-you-go system is not able to meet the obligations to a large number of pensioners and this is further exacerbated by unfavorable demographic trends. On the other hand, pensioners receive a low benefit despite high contributions to pension insurance. The greatest weaknesses of the public pension pillar include a high deficit that requires direct transfers from the state budget and an outstandingly low share of old-age pension beneficiaries compared to other pension beneficiaries. Furthermore, inadequate replacement rate and insufficient interest in voluntary pension savings, which is primarily caused by financial illiteracy, fail to alleviate poverty in old age. The original contribution to knowledge emerging from this doctorate is via the uniform evaluation of economic and social effects of the reforms of the Croatian pension system within a general equilibrium model, which made it possible to assess the optimal means of financing among the five proposed alternatives. Using an overlapping generations general equilibrium (OLG) model enables a comparison of reform implications that cannot be achieved using conventional econometric methods and models. Both macroeconomic and social implications were studied, taking into account the conflicting goals of sustainability and adequacy, to conclude which of the reforms can be regarded as the most favorable solution for the pension system and economic system. Additional contribution of this doctorate is reflected in the OLG model calibrated on Croatian data, laying the foundation for future empirical research of the Croatian pension system and comparable systems, as well as many other scientific fields. The methodology applied extends Diamond’s (1965) neoclassical OLG model, building an OLG framework that allows for the existence of several generations, the government, and a two-tier pension system. The model economy is populated by three overlapping generations to accurately depict the three-period life cycle of a representative Croat, who is entitled to an old-age pension at age 65 after 40 years of pensionable service, and who then receives a pension for approximately 20 years. Individuals pay contributions for both the first and second pension pillars, while the government, in addition to revenue from pension contributions, also taxes consumption and income from labor and capital, and issues government bonds. A representative firm produces a single good using aggregate labor and capital stock as inputs and the Cobb-Douglas production technology. The benchmark economy targets key Croatian macroeconomic data in 2019, where some parameters are calibrated and others are taken from previous research or match actual policy settings in the base year. Empirical implementation of the OLG model enables the evaluation of reform scenarios. This thesis examines the effects of the following five pension reforms: complete transition to the fully funded pension scheme (1), complete transition to the pay-as-you-go pension scheme (2), a reduction of total pension insurance contributions (3), an increase in the contribution rate for the first pension pillar (4) and switching to a pure defined benefit system (5a) or a defined contribution system (5b). The findings of the simulations can be summarized as follows: individual capitalized pension savings have a negative effect on the fiscal position of the economy, but lead to a growth in the accumulation of capital and aggregate savings, economic growth and welfare (1); strengthening the public pension system increases its sustainability, but its negative effect on a number of macroeconomic categories undermines the positive implications (2); a reduction of total pension contributions promotes the growth of consumption and gross private investment (3); on the contrary, increasing spending on the first pension pillar improves the fiscal position of the economy, but insufficiently to close the gap between pension expenditures and contributions with negative effects on welfare (4); the defined contribution system is more appropriate in an aging economy than the defined benefit system, but both alternatives are inferior compared to the Croatian public pension system established on pension points (5). The results of the thesis found clear support for implementing parametric reforms of the first pension pillar and reducing pension insurance contributions