OUR electoral system is proportional - but imperfectly so. Imperfectly, because it is designed to be proportional not at national, or even regional level, but only within constituencies. And while up to 1933 there were some eight and nine seat constituencies, and up to 1948 some seven seat constituencies, during the past half century our electoral areas have been limited to a maximum of five seats.
Now, clearly, with constituencies of this limited size, smaller political parties and Independents with minority support tend to be under represented in the Dail. For in a five seat constituency it is difficult for any individual or party to win a seat with less than 10 per cent of the vote, and in most elections few Independents or micro party candidates achieve such a vote.
As a result of this, many of these votes accrue to the eventual benefit of the larger parties, which thus almost always secure a higher proportion of seats than their share of the vote would warrant.
What determines the extent to which each of the two major parties benefits from this seat "bonus"?
In principle, the distribution of this "bonus" for the larger parties is determined by the later preferences of these voters, where they express such preferences by voting down the ballot paper. And changes in these later preferences between one election and the next could influence the final result as between the parties that win seats.
In practice, however, shifts in these later preferences of voters whose first choice of Independents or microparty candidates has failed to yield seats, do not seem to be a significant factor. Indeed, while I am open to correction on this point, I have in fact been unable to find evidence that in this election changes since 1992 in the pattern of preferences received from any source, including the Progressive Democrats, swung a single seat to Fianna Fail.
I know that this is contrary to the impression left by a number of commentators, and indeed Fianna Fail spokesmen, who in the last few days have drawn attention repeatedly to what they allege to be the benefits that Fianna Fail has secured from such an improved transfer pattern. And, in case I am mistaken in this, I shall be happy to listen to any arithmetic analysis of the voting pattern in a constituency that will demonstrate the contrary.
What can, and does, swing seats, however, is the way the two major parties "manage" their first preferences as between their candidates. This has been a major factor affecting the outcome of elections ever since Jack Lynch's decision in 1979 to put a stop to gerrymandering by appointing an independent commission to determine constituency boundaries a practice that all subsequent governments have followed, and which has now been enshrined in legislation.
The disappearance of that disreputable method of distorting the outcome of elections forced the two major parties to look for a legitimate way of maximising the number of seats produced by their first preference votes. How does this work?
If in some key constituencies a large party's candidates can be persuaded - or sometimes perhaps constrained - to agree that those voting for that party in different parts of the constituency be asked to cast their first preferences for one or other of the party's candidates, with a view to dividing the total party vote fairly evenly between them, then that party can sometimes gain an extra seat.
And this works particularly well if at the same time the other large party fails to restrain selfish attempts by one or other of its candidates to "hog" that party's vote with a view to topping the poll.
It was because Fine Gael successfully adopted this tactic in the November 1982 election while Fianna Fail, led by its leader in his own constituency, notably failed to do so - that the party I then led secured three or four extra seats on that occasion, which ensured our capacity to form a government with Labour which lasted for almost four and a half years.
In the 1992 election the microparties and Independents won 9.6 per cent of the vote - but instead of winning the same proportion of the seats, viz. 16, they secured only six, leaving 10 seats for the larger parties to win over and above those to which their own first preference vote entitled them.
But Fianna Fail on that occasion secured only three of these 10 "bonus" seats - instead of the six that would have been appropriate to its share of the large party vote. Indeed, on that occasion, Fianna Fail's poor vote strategy in the constituencies led, most unusually, to half of its potential six seat bonus being shared by two other smaller parties, the Progressive Democrats and Labour.
In this election, Fianna Fail has done better. First of all it has recovered from the Progressive Democrats the two bonus seats that the PDs had been allowed to get away with in 1992. And it has also taken three more Progressive Democrat seats which, on the basis of that party's overall national first preference vote in this election, the PDs might reasonably have been expected to hold.
In three of these five cases - Cork South Central, Limerick East and Waterford a principal factor in the loss of these PD seats to Fianna Fail was, of course, the disappearance of the Progressive Democrat TDs elected in 1992 - Pat Cox to Europe, Peadar Clohessy to retirement, and Martin Cullen to Fianna Fail. But in Dun Laoghaire and Cork South Central, vote management was the principal factor enabling Fianna Fail to win these two seats from the Progressive Democrats, although in the Cork constituency a vote swing from the PDs to Fianna Fail was a further factor.
The result of all this has been that instead of the 10 seats that she PDs won with their 4.7 per cent of the vote in 1992, this time, with precisely the same share of the vote, that party has won only four seats.
At the same time, the much increased number of votes that have been cast for small parties and Independents on this occasion expanded the number of bonus seats available for the larger parties from 10 to 15. For, if the representation of these individuals and small groups in the Dail were proportionate to the votes cast for them, the Green Party, Sinn Fein, microparties and Independents would have secured 25 seats between them while in fact they won only 10.
Now of these 15 "bonus" seats, Fine Gael's share of the total first preference vote broadly entitled it to six, and Fianna Fail's share entitled the party to nine. However, contrary to what Moire Geoghegan Quinn asserted in The Irish Times yesterday, Fine Gael's continued superior vote strategy won it eight of these seats, whereas Fianna Fail won only seven two less than its share.
Fianna Fail might, indeed, have got more rather than less than its share if its lead candidates in four constituencies in particular - Carlow Kilkenny, Longford Roscommon, Kerry South and Louth had in each case shared his votes with his fellow candidates. In other words, continuing deficiencies in Fianna Fail constituency strategy prevented it from securing the 79 seats it was due, and the 81 it might have obtained.
The fact that Fianna Fail, with precisely the same share of the vote as in 1992 secured an additional nine seats in this election does not seem to reflect an improvement in preferences passing to it from other parties, but was due rather to a combination of a largely chance reversal of the split of seats between Fianna Fail and the PDs, together with some - but not very much - improvement in Fianna Fail vote management.