All governments, every year, face three hard choices. They have to pull their ideas together into an annual public spending plan; a tax plan; and a Queen's Speech of proposed legislation to put to Parliament.
Let us start with spending and its mirror image, tax. Efficient government self-evidently depends on the ability to forecast public spending accurately; to choose spending priorities effectively; and to finance spending adequately. On all these scores, Britain has had a pretty chequered post-war history.
Flip back through decades of public expenditure reports and you will find that it has been depressingly rare for spending trends to match up precisely to a government's declared priorities. Something has long needed to be done to improve the government's strategic strike rate. The new Budget system was intended to do just that.
Three years ago changes were announced that in a more constitutionally minded country would have attracted a great deal more attention: a structural overhaul of the way government decides how to raise and spend money. The Chancellor abandoned the quaint practice of announcing a Budget that was not a Budget in the way most people understand the word. That famous little Budget box now bulges with proposals for spending money, not just for raising it. From 1993, tax and spending have been brought together, and announced in a single, late November statement. Less obviously, but even more significantly, the way decisions on spending are taken has been changed.
Spending is now reviewed, programme by programme, by a committee of senior ministers chaired by the Chancellor, which assembles in the Cabinet Office. This development is significant in itself. It is also a hugely important boost for the Cabinet committee structure that underpins our system of government.
Cabinet itself is usually - but not always - a formality. A body of 20- plus ministers cannot reasonably debate and reach detailed conclusions on issue after issue; but Cabinet committees can and do. My bet is that the spending committee, EDX, willbecome institutionalised, almost constitutionalised - and would do so even if the Labour Party were to win power. Tony Blair, even more than John Major, would need some such mechanism for corralling his team into agreement on public spending.
Nevertheless, the new system is likely to evolve because the present situation is not stable. One problem is that even if this new spending committee reduces the overload on the prime minister, it adds to the load on senior ministers. Last year's "round" of meetings must have absorbed something approaching 200 ministerial man-hours. The Treasury is having to learn to operate differently. Reform, once begun, has a habit of setting its own agenda.
A second problem is that the ministers assembled to act as jury over spending decisions have still had no part in the decisions over tax. The Treasury has clung jealously to its right to decide taxation unilaterally. It is unlikely to be able to maintain this barrier much longer. Sooner or later this side of the government's balance sheet is bound to become a matter of collective discussion, too. That may even lead the Treasury towards the still more open process of publishing a "Green Budget" of forecasts and tax proposals. That would entail abandoning the traditions of "purdah" and internal mystique that have surrounded the Budget process for far too long.
This change in the Budget-making system, however welcome, has made the parliamentary year extremely top-heavy. November is now crowded with big announcements: the legislative programme, tax changes and spending plans. By contrast, the spring is a tedious trek across the parliamentary plain, as one Bill after another passes from second reading to committee stage and out again. Sooner or later, government is likely to take another look at the parliamentary year - either moving the official start, or instituting some big set-piece "state of the nation" occasion to fill the "spring gap".
This brings me to the third key activity of government: how it makes law.A glance back at the Queen's Speeches of the post-war period illustrates nothing so much as the perpetual triumph of political hope over administrative experience. It is not merely that there has been "too much" legislation, or that it has so often had to be revised. But there have been some sorry examples of rapid revision in recent history: the community charge, the 1991 Criminal Justice Act, the Child Support Act - all had to be changed almost before the ink on the stature book was dry. None of this breeds great confidence in the law-making process.
Nor is it enough simply to focus on the distinctly patchy revising skills of Parliament. Where government does have a strong interest of its own is in avoiding the embarrassment of revision and repeal. For that, the vital question is whether the heart of government - the unsung Cabinet committee structure - can reinforce its own filters for legislation, wrenching more control from departments or imposing greater supervision on "their" legislation.
More open Budget-making, more open law-making, a stronger centre of gravity to Cabinet government: to achieve these we need no glamorous new constitutional architecture, merely changes in procedure. But is that enough? An equally strong line of attack today is focused on the accountability of government - and to judge that we must step outside the machine.
It is ironic that it should be the government's effort to decentralise and disperse power beyond the bureaucrats and "experts" that should have aroused this attack. Debate is healthy, but what is needed is to broaden the argument beyond electoral accountability - the voter's intermittent right to hire and fire - to the "new accountability" that the government has been seeking to develop throughout the public services. Through published standards, information about whether they are met, real choices of schools or hospital treatment, systems of redress, this new accountability can operate not once every few years at the ballot box but day by day, direct from the providers of services to the people who use those services.
The change in the way public services are being delivered is still patchy, but profound. It is part of a final paradox: that while Britain's constitution has survived remarkably unchanged, the role of the state in Britain has been changing very fast. Agencies, regulators, self-governing bodies, inspectors - all have to settle into new relationships with each other, the state and the citizen. Throughout the public sector connections between purchaser and provider, player and whistle-blower, rule-maker and contractor, are being exposed to view. The smooth stone Whitehall facade has become more like the Lloyd's building in the City: all the shiny working parts are on show.
This new transparency is a powerful tool, not only for those who run government, but for those who judge it or seek access to it. There is rich terrain here for the reformer - both inside and outside government - who is able to cross boundaries, defining and refining the new pluralism of the modern British state.
The author was head of John Major's policy unit from 1990 to1995. This is an edited version of an article in `Prospect', a new monthly magazine, available from Thursday, pounds 3.50.Reuse content