School admissions will be at the heart of the imminent white paper. Until now, Labour has swerved around the knottiest issues, such as pupil selection at grammar schools.

This time, though, the policy on admissions will be the acid test of the Education Secretary’s boldness, or otherwise, in pursuing her stated aim of improving the access of children from deprived backgrounds to the best schools. Ruth Kelly has said Labour’s education policies should be about improving social mobility. How far will she go to overcome the perceived disadvantages of the poor, the ill-informed, and the non-religious in school admissions?

We ought to expect something radical. At the Labour party conference, both the Prime Minister and Ruth Kelly stressed the current inequity. As Ms Kelly said, access to some schools was “open only to those who could afford to buy an expensive house next to a good school”. This prompted speculation that the white paper would enforce new methods of deciding admissions - banding or even lotteries - which would reduce the advantages of the middle classes.

But last week, the Prime Minister told the House of Commons that “you cannot introduce choice simply by a choice mechanism”. So the main strategy for improving choice in the inner cities will not be positive discrimination through admissions policies but rather a continuation of attempts to improve failing schools or to replace them with academies.

The white paper will focus on the supply-side constraints of school choice by encouraging good schools to expand and helping parents to open schools.

Yet the debate over admissions will not go away. Until we reach the utopia where every school is a good one, controversy will continue to surround the way over-subscribed schools ration places.

Recent research from the Sutton Trust gave the Government a last-minute reminder of the, mostly unintentional, social inequities produced by admissions systems at the top 200 schools in the GCSE performance tables.

It found that only 3 per cent of pupils at these schools were from families eligible for free school meals. By contrast, in the postcode neighbourhood of these schools, the eligible proportion was 12 per cent, four times higher than the schools’ intake. This gap was as wide for comprehensives as for grammar schools. Sir Peter Lampl, chairman of the Sutton Trust, claimed the best state schools were thus “socially selective”, whether or not they were academically selective.

School admissions are a maze. Each of the 150 local authorities can set its own admissions policies. So too can every single academy, city technology college, foundation school and voluntary-aided school.

Since 1944, admissions policies have taken some account of parental preferences. Yet, when negotiating these mazes, some parents are more equal than others. Numerous factors bring advantages: the ability to buy an expensive house, membership of a church, car ownership, access to information on school performance, or simply being able to pay for extra tuition, costly uniforms or public transport.

The admissions maze has also become more complicated as a result of the policy of school diversity pursued since 1997. As in the trendiest bistros, there is now a bewildering menu on offer to parents in many parts of the country.

Specialist schools seem to come in 57 varieties. Sikh, Muslim and Jewish schools have been added to the traditional Anglican and Catholic fare.

There are foundation schools, grammar schools, semi-selective schools, academies, and city technology colleges. The diversity policy is based on the notion that parents can choose between schools. Yet this is where the menu analogy fails; in a good restaurant everyone can choose the special of the day; in school admissions not everyone can get into the most popular school.

Last year, the Commons education select committee expressed concern that the balance of power was shifting away from parents choosing schools, towards schools choosing pupils. It said admissions arrangements at too many schools failed to “command the confidence of parents” and that the process was often “a frustrating and time-consuming cause of much distress” for many families. It concluded that “fairness in public policy should not be a matter of luck but a matter of course”.

Since the select committee reported, the Government has produced a new draft school admissions code of practice. This is due to be implemented early next year, replacing the 2003 code.

Yet, according to Chris Waterman, who represents England’s local education authority leaders and has written a detailed analysis of the 2005 draft (Sins of Admission, IRIS), the new code is “weaker” than the 2003 version in transparency and fairness. In a foreword, Barry Sheerman, the chairman of the select committee, describes the new code as “a very pale imitation of what is needed”. The biggest problem, according to Mr Waterman, is that the law still only requires schools and authorities to “have regard to” the code. He believes it should be binding.

The new code attempts to deal with admissions criteria that have caused grievances. For example, it says schools should not use a first-come-first-served method that might favour those parents who are plugged in to the middle-class grapevine. It warns against, but does not condemn, admissions criteria that favour those who make a school their “first preference”. It also deals with issues arising from the London Oratory school (chosen by the Blairs) which uses interviews as part of its admissions procedure. The draft code says interviews with either children or parents are “poor practice”.

It similarly deprecates any attempt to admit pupils on the basis that their families are “more religious than others”. However, this censure does not force schools to stop using these methods.

The forthcoming white paper is expected to endorse “banding”. This involves setting a test for all applicants who are then placed into ability bands.

The school must then admit pupils in equal proportions from each band.

Banding has been used in the past, notably by the defunct Inner London Education Authority, to ensure a genuinely “comprehensive” intake. The current law on banding requires that applicants represent the ability range of all the applicants, not of a local education authority or national average. However, where selective schools exist, banding based on the ability spread of applicants is unlikely to bring a genuinely comprehensive intake. So the Government is investigating whether to base banding on the ability range of all pupils locally. Banding is likely to find favour in the white paper because it levels up the chances of pupils from families less adept at playing the system. Sir Cyril Taylor, who chairs the Specialist Schools and Academies Trust, supports banding and, if the white paper endorses it, would recommend it to his member schools.

However, suggestions that the Government will introduce mandatory banding are wide of the mark. It might deliver Ruth Kelly’s aim of greater social mobility, but it would frighten the middle classes and would run counter to the Government’s stated belief in school autonomy.

“Random allocation”, or in effect, names out of a hat, is another method which aims to level up the chances of pupils from poorer backgrounds. One of the new academies, Haberdashers’ Aske’s Hatcham in Lewisham, is using this method. It had 2,500 pupils chasing just over 200 places. For the first time, the 2005 draft code endorses “random allocation”. Sir Cyril has seen it employed to effect in the United States and has “some sympathy” with it but believes it can be “disheartening” for everyone involved.

Although the white paper will address school-choice mechanisms, it is likely to focus more on the constraints that impede choice. The cost of travelling to a better school could be overcome by bigger transport subsidies.

Lack of knowledge or information about the choices available can be another constraint. Research commissioned by the Government found that families where the mother had a university degree were five times more likely than other families to use sources of information, such as Ofsted reports and performance data, to inform school-choice decisions.

In other words, a consumer-choice system of admissions inevitably favours the children from educated homes, thus perpetuating disadvantage. One way the Government might deal with this is to borrow an idea from the National Health Service where choice advisers help patients to find the best treatment available.

In rural areas, much of the fuss over admissions may seem like a London-centric (even Islington-centric) concern. Yet over the past few years, around 10 per cent of all secondary-school applications have resulted in appeals. This is up from 6 per cent a decade ago.

Appeals are only the tip of the iceberg; many disgruntled parents see appeals as futile. Department for Education and Skills research suggests that only about 85 per cent of parents were offered a place in their favourite school. In London, this fell to 70 per cent.

So, politically, there could be gains to be made from a new national policy on admissions; but this is a risky game. Will Ruth Kelly stick her neck out?

Mike Baker is education correspondent of BBC News

Follow the code

Admission arrangements should aim to ensure that:

* Parents’ preferences are met where possible;

* Admissions criteria are clear, fair and objective;

* Local parents gain a place at the local school unless there is good reason to prioritise others;

* Parents have easy access to admissions information.

Acceptable oversubscription criteria:

* Catchment areas;

* Siblings;

* Feeder schools;

* Distance from next nearest school;

* Public transport;

* Religious affiliation (faith schools);

* Academic selection (grammar schools).

Poor practice:

* Interviewing pupils or parents;

* Favouring families which are more religious than others;

* Sibling policy which disadvantages ethnic groups recently arrived in the area;

* Aptitude tests which test ability.

Source: School admissions code of practice, draft 2005