Economia Politica. Rivista di teoria e analisi Economia Politica. Rivista di teoria e analisi
Non-Technical Abstract
Pensions present a common problem in Europe and the reforms being carried out at the end of this century constitute productive 'national experiments' which can one day merge into one 'sole European model'. My principal task here is just that of commenting on the recent pension reform in Italy passed by the Dini Government.
The reform was a long and difficult undertaking which involved both government and parliament for a good part of 1995, and which did not fail to attract a certain international interest. The initial objective of the government was to resolve some important problems left over from the 1992 reform. In particular, penalisation mechanisms were necessary to discourage the growing recurrence of early retirement. Altogether unexpectedly, and because of a special chain of political events, whose probability a priori was very low, a truly new and radical reform came into being. This reform profoundly altered the system, but did not change its unfunded, pay as you go nature. In previous years, I had explained the theoretical guidlines for this reform on more than one academic occasion as well as in a study carried out in 1993 on behalf of the General Accounting Office.
I proposed a pension system which, while remaining an unfunded, pay-as-you-go system, was ensured against 'long run' risks of 'permanent' deficit almost like a funded one. The concept originated with a result known to economists for some time: namely, that a balanced budget, unfunded pension system can remunerate the contributions according to a rate of return which is equal to the sum of the growth rate of the average salary and of the employment growth rate. In sum, a balanced budget, unfunded pension system is guaranteed against deficit by respect for this 'golden rule': namely, that the rate of return be equal to the growth rate of total taxable income. The deficiencies of our present pension schemes mostly derive from the chronic delay with which governments are able to impose respect for this rule.
I proposed, therefore, that the formula for calculation of pensions based on the last salary (or last several salaries) be abandoned, and that the same formula for defined-contribution, funded schemes be adopted. The rate of return, however, must be that indicated by our golden rule rather than by the interest rate normally yielded by financial investment. In this way, the system would adapt automatically and instantaneously to demographic and economic shocks, quite apart from the will and capability of government intervention.
A second advantage would be obtained under the profile of 'equity in space'. In fact, the rate of return would be unique at each point in time for every participant in the system, whether working people or retirees, just like the unique return offered by an insurance company.
I realized that this advantage could be easily taken for a disadvantage, considering the widespread idea that a pension sytem must contain elements of 'solidariety', and thus must allow for transfers of income from the well-to-do to the needy. In order to prevent this criticism, I determined to analyse the many hidden transfers which were implicit in the formula at that time used for calculating pensions based on the last five salaries.
As expected, the analysis showed enormous transfers, but unexpectedly revealed as well, that these transfers went in the exact opposite direction to that suggested by the normal criteria of solidariety. For example, it emerged that transfers occured from a flat career (i.e. a constant, salaried career) to a dynamic one (i.e. an increasing salaried career), from a fragmented working life (frequently interrupted) to a continuous one, from a brief working life (which compelled posponement of retirement until full retirement age was reached) to a lengthy one which permitted early retirement.
Thus, the analysis of implicit returns revealed that the last salary formula permitted serious 'injustices' by transferring income from the less to the more well-to-do, and in particular from blue collar workers to white collar workers and from both these categories to upper and middle management. A 'liberal' formula would then be preferable: this would restore to each retiree his/her pension which is earned by means of contributions individually paid and by means of an interest rate which is strictly equal for everyone.
It was, in fact, this denunciation of inequities inherent in the traditional formula which in 1994 claimed the attention of the political left and the unions. Shortly and rather unexpectedly, the left and the unions would become important supporters for the technocrat government of President Dini. And all this allowed my proposal to become a matter for political discussion.
Not only could the new formula permit financial balance of the system and equity for participants, but it also allowed for the non-problematic preservation of the flexible age of retirement which is so deep-rooted in the Italian social security tradition. In other words, the new formula permitted the resolution of the problems of financial unsustainability caused by early retirement. In the final version, the reform permitted leaving work at between 57 and 65 years of age; however, the amount of the pension increases in proportion to the length of time that the age of retirement is postponed: the working person who retires later has the right to a higher yearly income because his life expectancy is less. The increase in the average life expectancy in the next few decades will lead to a reduction in pensions for all retirement ages because the number of years the pension can be enjoyed will increase.
In a system like the one outlined, the contribution rate is no longer an endogenous variable which is determined by a preselected ratio between pension and salary and by the demographic ratio between pensioners and working people. This is, instead an exogenous variable which distributes wages between current and future consumption. The slackening of demographic expansion will compel the acceptance of a contraction in current consumption (disposable salary) in order to leave future consumption unvaried (pensions), or rather to renounce future consumption in order to leave current consumption unvaried. The contributive rate will be the instrument by which the desidered choice will be accomplished, while social equity and the financial balance of the system will be in any case ensured by respect for the golden rule.
Should we then be satisfied with the work accomplished? Have we truly constructed a system where one needs only to decide the distribution of income between the present and the future, while spatial equity and financial balance are ensured by an 'invisible hand' or, if you prefer, by an 'automatic pilot' which continually adjusts pensions to the growth rate of the average salary and of employment as well as to life expectancy?
Unfortunately, many criticisms cuold be made of the Dini reform. Some would be rather technical and would refer to particular aspects of the reform. The most important criticism, which to me appears decisive, lies in the 'non credibility' of the mechanism chosen to index pensions.
We have said that financial balance and spatial equity are ensured by respect for the golden rule on the basis of which one must remunerate contributions according to the growth rate of overall wages. Even with this constraint, there is the liberty of combining in many different ways the first yearly pension income with indexing. A different pension 'temporal profile' corresponds to each possible combination. There is a 'flat' profile when the yearly pension payment remains the same in real terms, and there are exponential profiles when the pension grows at a constant rate. Naturally, the first yearly pension payment is as high as the rate of real indexing is low and it is at maximum when this rate is nil, or rather when the profile is flat. In regard to exponential profiles, the 'normal' case is when the pension grows at a rate which is exactly equal to the rate of return suggested by the golden rule. Here then, the first annual pension payment can be calculated by dividing the sum of capitalized contributions by life expectancy.
Of all possible profiles, the reform chose the flat profile, so as to permit the first yearly pension payment to be maximized.
This choice generates the phenomenon of the so-called 'vintage' pension and thus, the diversification of pension by year of commencement. Therefore, this selection presages periodic, real adjustments of the oldest pensions which, as they are not provided for in the reform, will certainly sneak in by the back door. Policy makers will certainly not be able to resist such a popular demand which is also perceived as socially just. Older pensioners will be impoverished not only in respect to new pensioners, but also in respect to all other income. It must be remembered that examples of a systematic lack of real, automatic or discretionary adjustments do not exist on an international scale.
Summing up, the promise to renounce real indexing of pensions appears to be a 'sham' used in order to make the first yearly pension payment more generous, but which will be impossible to maintain. Equalization will be necessary and will unhinge once again the financial balance of the system because the rate of return will be pushed beyond the rate of growth of overall wages and the golden rule will be violated. As well, the individual rate of return will again be diversified by favouring those working people whose retirement period might be characterized by more frequent or more generous adjustments.
It must be understood soon that a simple retouching will not be sufficient since the introduction of a real indexing mechanism for pensions calls for a sharp decrease in the first yearly payment of about 15% to 20%.
SANDRO GRONCHI is professor of economics at the Università degli Studi di Roma "La Sapienza", Facoltà di Scienze Politiche, Dipartimento di Teoria Economica e Metodi Quantitativi, Piazzale A.Moro 5, 00185 Roma
S.Gronchi@sunset.spol.uniroma1.it
Go back to the Home Page 
Go back to the Index
Redazione: econpol@spbo.unibo.it
Webmaster: lrossi@spbo.unibo.it