Employee representation in the public administration simplified by reform

Download article in original language : IT9806229FIT.DOC

Negotiations over the renewal of Italy's public administration national collective agreements are underway in mid-1998. The talks are the first occasion on which new regulations on trade union representativeness in the sector have been applied. The rules have helped considerably in simplifying the composition of worker representation at the negotiating table. This article reviews the selective effects of the new rules, using a new source of data.

Negotiations began in mid-1998 between the public sector bargaining agency, Aran (IT9711217F), and the trade unions over renewal of the national collective agreements for the public administration (all of which expired on 31 December 1997). The talks are the first time in which the new regulations governing trade union representativeness in the sector, introduced by legislative decree 396/1997 (IT9709311F) have been applied.

The fragmentation of representation in the public sector

The typical features of trade unionism in the Italian public sector are a higher union density than in the private sector, and the presence of a large number of organisations, often of small or extremely small size. The three large trade union confederations (Cgil, Cisl and Uil) are present to a considerably lesser extent than in the private sector and account for less that 70% of public sector union membership - which is far from being a monopoly.

This fragmentation is the result of both the political-ideological cleavages typical of Italian trade unionism - divisions strengthened in the public sector by the more direct involvement of the political parties in industrial relations - and by the presence of several organisational criteria unknown in the private sector. In fact, besides industrial and territorial criteria, which still predominate, numerous trade unions are based on craft or occupational criteria, or indeed factors to do with the public administration body to which workers belong or with their "position". This is indicative of a labour market characterised by a large number of professional workers, jealous of their specific character, and endowed with considerable capacity for disruption.

A new source of data

Both the features mentioned above are confirmed by figures recently published by the Department of the Public Function (Dipartimento della Funzione Pubblica). These figures concern the authorisations signed by public sector workers for union fees to be deducted from their pay (check-offs). This information is therefore more reliable in assessing union membership than the enrolments declared by the unions, the alleged totals of which have sometimes been greater than the number of employees.

As far as unionisation is concerned, out of a total of just under 3 million "contractualised" public-sector employees (ie those whose employment relationships are defined by collective agreement) around 45% are unionised, a value which is several percentage points higher than in the private sector. Just over 30% belong to Cgil, Cisl and Uil, and the remaining 14% to the so-called "autonomous" unions.

This new source of data also yields especially rich and sometimes surprising information on union fragmentation.

The "contractualised" public sector is divided into eight branches - ministries, non-economic public bodies (ie public social security), regional and local authorities, autonomous state bodies, health, research institutes, education and universities - and three staffing categories (usually called "areas") - non-managerial employees (approximately 2.7 million), managers (55,000), and doctors and veterinary surgeons (105,000). According to the Department figures (see table below), a total of fully 714 trade union organisations "of first affiliation" operate in these eight branches and three categories. Even after it is taken into account that many of these unions belong to confederations or second-level organisations, which may cover several branches or areas, the figure is still 412 organisations. This overcrowding, and the inter-union rivalry that results from it, is not uniform in all branches and areas, and it is on the whole much greater among managers than among non-managerial personnel, if their respective numbers are taken into account.

* As defined under provisional rules (see main text).

Source: Department of the Public Function.

There are further significant indicators of fragmentation:

• the unions not affiliated to the three large confederations are on average very small in size (with union check-off totals ranging from a few dozen to a few hundred);
• more than half (52%) of the unions of first affiliation, and more than 46% of the confederations, account for less than 0.1% each of union check-offs in their respective branch or area - a figure which is 40 times less than the minimum representativeness threshold fixed for the first year of application of legislative decree 396/97 (see below). In particular, these figures reach almost 70% in relation to non-managerial personnel in the three largest branches (local authorities, health and education); and
• fully 31% of the unions of first affiliation, and 28% of the confederations, do not even have more than 10 authorisations for union check-offs to their name (and 10%-11% only receive one such authorisation!).

The selective effects of the new law

The new law recognises as representative - and therefore admits to national-level collective bargaining - those trade union organisations with 5% representativeness in their branch or area. Representativeness is calculated as the average of a union's membership figure - identified by its percentage of check-offs out of the total number issued in the respective branch or area - and its electoral results - the percentage of votes obtained in the elections for the recently-created public sector "unitary trade union representation bodies", the Rsu s (IT9711138N). However, as a transitional measure (until the Rsu elections are held), the decree stipulates that only unions receiving at least 4% of the total number of check-off authorisations in their branch or area, as well as the confederations to which these unions are affiliated, may be admitted to national-level bargaining.

The application of this criterion has simplified matters greatly, especially in certain branches. Among non-managerial personnel in local authorities, for example, only three unions out of 107 exceed the representativeness threshold; in health (non-managerial and non-medical personnel) four out of 83; and in schools five out of 62. The selective effect is proportionally greater in the area of non-managerial personnel, where the number of unions falls from 459 to 35, than it is among doctors and, especially, managers. It is also particularly pronounced among the confederations, many of which were allowed by the previously more permissive legislation to take part in national-level bargaining - even if there were no representative unions affiliated to them in the branch or area concerned.

Commentary

The new rules introduced by legislative decree 396/97 finally defined clear criteria for union representativeness, and therefore for access to collective bargaining and union rights, filling a gap which had been opened by the results of the referenda of June 1995 on representativeness (these resulted in the repeal of 1993 regulations providing for an agreed definition of union representativeness in the public sector). Union delegations in the current public administration negotiations are simplified in comparison with those of the past, especially in terms of the number of confederations. This simplification is taking place despite a situation that remains greatly fragmented, and should be even more evident when the final criterion of 5% (as the average of check-offs and votes in the Rsu elections) replaces the transitory one now in place, which is slightly less selective (4% of check-offs alone).

However, it is likely that the new law will induce some of the smaller unions to merge so that they can achieve the representation threshold. This process was already apparent in the 1980s after the first laws on such matters were passed (laws 93/1983 and 395/1988), but this process should now involve unions belonging to the same branch or bargaining area. Even though this could represent merely an easy way out to avoid the new rules on numerical grounds, on the other hand the possibility cannot be ruled out that it could also lead to another sort of simplification, since the various organisations would be obliged to abandon a strictly "particularistic" attitude to representation, in favour of a more encompassing and more responsible one. (Lorenzo Bordogna, University of Brescia).