Lord Jenkins's report on electoral reform was published yesterday, but it simply confirms that there is no such thing as a perfect system, writes David Butler

Margaret Thatcher did not like using Royal Commissions or departmental committees to find answers to difficult policy questions. Tony Blair does. And he has been well served by his first two major efforts.

Lord Neill and his nine colleagues on the Committee on Standards in Public Life took 11 months to sort through the jungle of problems associated with the finance of parties. Two weeks ago they came up with answers that have won general acclaim, even if they offer considerable embarrassment to both the main parties.

Now Lord Jenkins and his four colleagues, also in a mere 11 months, have produced a remarkable contribution to the study of electoral systems and, perhaps, to the dilemmas inherent in the alliance between Tony Blair and Paddy Ashdown.

É«ºÐÖ±²¥

The prime minister gave Roy Jenkins and his notably civilised team an impossible remit. They were charged with finding an alternative to Britain's long-established first-past-the-post electoral system, an alternative that, in fulfilment of Labour's manifesto promise, could be put to the people in a referendum. But they were given four "requirements" that, as the commissioners modestly observed, were "not entirely compatible". They were: (1) broad proportionality; (2) the need for stable government; (3) an extension of voter choice; and (4) the maintenance of a link between MPs and geographical areas.

Broad proportionality is plainly inimical to clear single party government (since no party has won 50 per cent of the vote since the war), while increased voter choice must involve the opportunity to vote effectively for smaller parties. Moreover, any move towards proportionality necessitates at least some enlarged constituencies and that must attenuate any territorial link between members and constituents.

É«ºÐÖ±²¥

But the commissioners have tackled their impossible task with great skill and enthusiasm. They have produced a report that bears the stamp of Roy Jenkins's lucidity and style in every sentence. It should win a prize in any Ernest Gowers Plain English competition and it contains several anthologisable one-liners. It gives concrete, and at times indiscreet, examples. It makes plain the essential impossibility of devising a perfect electoral system.

Anyone charged with devising how to turn votes into seats and thereby into MPs and governments faces the challenge of finding the least bad system, knowing that it will only be a passable answer for a particular country at a particular point in time, fully aware that the right solution today may be wrong a generation hence and that it would certainly be wrong for another country.

The commission was given the challenge of coming out in favour of a single system because those who want change understand the principle of "divide and rule". Every one of the myriad possible systems has its snags, as well as its zealous advocates. Enthusiasts for change, aware of the awful example of the New Zealand referendum of 1992 (where the voters were asked to choose from a cafeteria list between four systems) accepted that the problem must be reduced to a simple, agreed alternative.

Roy Jenkins and his colleagues set about finding that alternative in an impressive way. They spent six months collecting evidence about electoral systems before even discussing the direction they were going to take. They travelled to Germany and to Australasia to see how other systems worked in practice. They commissioned research on the likely outcome of different methods. They asked a group of academics to answer a long list of technical questions. I chaired the group last June, and at no point did we discuss what system should be adopted nor were we given any clue as to where the commission was directing its focus. Our answers to the questions put to us (which were remarkably consensual) are recorded in an appendix to the report.

É«ºÐÖ±²¥

It seems quite clear that the general approach, the research methods and the style of the report should be applauded. But do we want the British people to vote in support of the recommendation?

One objection should be dismissed. The proposed system seems complicated. It combines the alternative vote in single member constituencies with a semi-proportional top-up in small regions. This is to be presented to voters who, since Labour came to power, will have seen Parliament legislate for proportional representation with party lists for the European parliament; for variants of the German Additional Member system for the Scottish and Welsh legislatures; for the Single Transferable Vote for Northern Ireland; and, probably, for the Supplementary Vote for the London Mayor, with first-past-the-post surviving only in its least defensible territory, local government.

Although there is ample material for satire here, most of the satire will be misplaced. The complications of the systems are dealt with by the professional counters. Ordinary voters, habituated to football pools and lotteries, will have little difficulty in expressing preferences between parties, putting two crosses on a ballot-paper or numbering candidates in rank order. Their counterparts in Ireland, Belgium and Australia cope easily enough with such tasks.

But has the commission squared the circle? It has substantially saved single-member constituencies. It does enhance the opportunities for voters to choose smaller parties without opening the door to a multitude of very small parties. But when it comes to combining broad proportionality with stable government, the jury must be out. In the great majority of elections since the 1960s, the new scheme would have denied the country a single-party majority. (It is possible to argue with the commission's arithmetic on this point.) Of course, many would applaud such an outcome. Moreover, coalition government (or minority government with minor party 'understandings') is not necessarily unstable. But it is very different from what we have known. If the Jenkins proposals go to a referendum in the next year or two (which is doubtful), if they get a 'Yes' vote (which is doubtful), and if they are then implemented (which, at the least, will take time), the rules of the game for cabinet politics and parliamentary politics and electoral politics will be very different from what we have always known. Meanwhile, those of us who have to teach or study politics will have great fun with this most readable of electoral treatise added to our bibliographies.

É«ºÐÖ±²¥

David Butler is emeritus fellow of Nuffield College, Oxford.