Statistics and Registration Service Bill

My Lords, as this debate has demonstrated, statistics matter, because they are central to the management of public policy by government and, just as importantly, they inform the citizen about what is happening in almost every aspect of life in our country. In some cases, of which crime statistics are perhaps the clearest example, published statistics directly affect an individual’s sense of well-being. As the debate has demonstrated, all is not well with the management of official statistics in the UK. In its survey, the Office for National Statistics found that only 17 per cent of the population believe that government statistics are produced without political interference and a mere14 per cent think that they are honest. Those are almost incredible figures. They reflect extremely badly both on this Government, who have a particularly disreputable track record in their use and abuse of statistics, and on our system of government over a number of decades. The Bill is a once-in-a-generation opportunity to put in place a framework that can begin to restore faith in government statistics, and we welcome it in principle. However, it is far from clear that it will achieve its aim. Virtually all noble Lords who have spoken in this debate have many concerns. They are shared by the professional statistics community, including the Statistics Commission, the Royal Statistical Society and distinguished former chief statisticians from a number of Commonwealth countries. A number of concerns have been expressed about the Bill. The House has made it clear that it wants to concentrate on four concerns in its detailed consideration of the Bill and that there is a fifth concern that the Bill cannot adequately address. The first concern relates to where responsibility for statistics should lie. The Bill places it with the Treasury. Like the majority of noble Lords who have spoken, we would like to see it shifted to the Cabinet Office, which seems logical. Within the terms of the Government’s thinking, if they are about to establish a central hub for the release of statistics, it would be bizarre to place it with the Treasury; the logical place would surely be the Cabinet Office. The principal argument for placing responsibility with the Cabinet Office is that the buck for maintaining the integrity of the statistical system should rest with the Prime Minister not the Chancellor. Under the current Chancellor, Treasury writ has run across much of the domestic public policy-making agenda, but that will change when he moves next door. It is also unusual and, in our view, unadvisable for the Chancellor to have quite such a wide remit. As a general principle, the Prime Minister not the Chancellor must be best placed to deal with abuses of official statistics at ministerial level. To put it crudely, the Prime Minister can sack a recalcitrant or misbehaving Minister, and even this Chancellor has not had that power. There is also the question of resources. Although it could be argued that the Treasury is best positioned to ensure that the statistical service has adequate resources, recent experience suggests that it may need protecting against a rather macho cost-cutting Treasury, against which, as the Bill is drafted, it would have little appeal. This is not a theoretical concern. As has been pointed out, the current resources of the statistical service are far from satisfactory. The ONS is moving to Newport and will undergo a further reduction in staff. Thestaff is already down from 1,000 to 600, and a further 250 job cuts are in the pipeline. The move to Newport has led to a third of the staff resigning, retiring early or taking redundancy, and I believe that, to date, only 40 of the 600 staff due to move have accepted the offer of relocation. This is already having a major impact on the quality of statistics produced. The ONS has already announced that this year’s Blue Book on the national accounts will not be published in full and that the annual statement of the UK’s balance of payments will include ““less detail than normal””. Can the Minister give us the latest picture on the movement of staff to Newport and say whether the Treasury is satisfied that the move and the staff reductions will not seriously impact on the ability of the ONS to meet its ongoing requirements? Given the current staffing shortages and the further cuts planned, I would be grateful if the Minister could explain what the Chancellor meant when he said in the Red Book that the ONS was establishing, "““a full regional statistical presence””," across the English regions. Is that to be one man, or woman, and a dog, or will there be adequate staff to produce the expanded range of regional statistics promised? The Minister said that we need not worry about funding for the statistical service because its decision-making is dealt with separately, and it has a budget for five years. So far, so good, but the bad news is that it appears that the budget is inadequate, so the budget for five years is inadequate. I fear that the assurances that the Minister gave about the budget-setting procedure do not reassure us. The second concern, expressed in virtually every speech, relates to the pre-release procedures. It is probably the area that causes the greatest concern. There is a growing practice of Ministers and advisers using the long pre-release period to massage how the figures are presented. This, more than anything else, has led to the fall in trust in government statistics. There was a long discussion of this issue in another place, but there was no satisfactory conclusion. The Government propose to deal with the code that covers pre-release in secondary legislation. Unfortunately, it has not been published, even in draft, so we do not know what it is likely to contain. However, we know that Ministers consider that 40.5 hours is an appropriate period for Ministers to have access to sensitive statistics before they are published. That is a bizarrely precise period; it is certainly far too long. It is very long by international standards. Even in the US, on which the Government have placed great store in debates in another place, the maximum notice for a very limited range of statistics is overnight. Even then, access to those statistics is much more tightly constrained than in the UK. The Government have said that the regulations that they propose to introduce will be reviewed after 12 months, so if they are found to be inadequate, they can be changed. That will not do. There is no reason why the regulatory framework cannot be got right from the start. Some noble Lords have suggested what the pre-release procedure and period should be. My colleagues in another place suggested that two hours should be adequate in virtually every case. We will have the chance to discuss the various options in Committee, but the House is clearly already agreed that the current provisions and the plans of the Government in this area are inadequate. I wish to mention two aspects of this House’s scrutiny of the Bill. First, the regulations to which I have just referred will come to Parliament as affirmative resolution statutory instruments. I think that the Minister and the noble Lord, Lord Desai, said that that was tremendous because we would then be able to discuss them. It is nothing of the sort. We all know that any secondary legislation is rubber-stamped by this House and another place. We cannot amend it. It causes a near constitutional crisis if we throw it out; therefore, the offer of affirmative resolution procedure, although it at least allows people more time to look at the regulations than the negative resolution procedure, is no reassurance that Parliament has a real say in what they contain. The second aspect of parliamentary scrutiny raised by a number of noble Lords—an aspect that the Bill does not and probably cannot cover—is how Parliament will scrutinise the new arrangements. The Bill simply says that reports will be made to Parliament. So far as I am aware, the Government have not expressed a view about how that might then be dealt with. Often, when we raise this kind of question, the Ministers say, ““This is for the House””. Again that is a sophistry. It would be of great help to the House if the Minister this afternoon could give us some idea how the Government see parliamentary scrutiny being exercised. The next concern relates to the difference between national and official statistics—where the line between them is and who draws it. This matters because the current boundaries clearly are not satisfactory. Examples have been given, such as the fact that quarterly NHS waiting-list figures are national statistics and monthly ones are not. Many Home Office statistics, in which trust is particularly lacking at the moment, are not even national statistics at all. I suspect that few people outside Parliament, and, indeed, most people inside Parliament, are aware that there is a distinction between the two sets of statistics. As the Minister pointed out in another place, quite correctly, there is obviously a great difference between the major issues such as census data, employment or balance of payment figures and the many detailed sets of statistics that have a narrow focus. My initial view was that we should simply categorise everything as a national statistic and be done with it. I am still quite tempted by that approach, as, I think, are the noble Lords, Lord Turnbull and Lord Moser. However, if there is to be a distinction, we need to be clear on where the decision-making on the boundary should lie. The view of Ministers is charmingly frank. Speaking on Report in another place, the Financial Secretary to the Treasury said of Ministers: "““We are responsible for making policy and, as such, we are arguably best placed to know which statistics are most critical to the development, delivery and evaluation of the policies for which we are responsible and accountable””.—[Official Report, Commons, 13/3/07; col. 196.]" That statement has the advantage of clarity; however, it is a mistaken principle that Ministers should have a veto on which statistics qualify for the enhanced scrutiny by the National Statistician and his staff that national statistics attract. We have concerns about the role of the National Statistician vis-à-vis the board. This is an area of major contention, but it is arguably the most difficult to get right under the proposed structure in the Bill. The issue is the distinction between executive and oversight authority. It is very important that the board does not attempt to second-guess the National Statistician on statistical production issues. Let us take the example of deciding, in cases like Network Rail, whether to designate a body’s liabilities as on or off the government balance sheet. The decision in that case is a technical one, in theory at least, that the National Statistician should take. The role of the board in such a case should be to satisfy itself that the process for making such a judgment has been properly applied, that there has been no undue political pressure or interference, that the code of practice has been adhered to and that the outcome has then been adequately communicated to the public. Noble Lords can see in that case that the board has a crucial but different role from the National Statistician. That difference has not been made clear enough in the Bill. While we agree with the noble Lord, Lord Turnbull, that we should retain the unitary status of the board and the National Statistician and his staff, we should try to clarify this area. This debate has demonstrated that there are serious concerns about the Bill but that there is a pretty clear consensus about what they are. They are relatively few in number and can be dealt with without a fundamental rewriting of the Bill. I was very encouraged at the start of our debate to see the Financial Secretary to the Treasury here to listen to the speeches of the noble Lords, Lord Jenkin and Lord Moser. I cherish the hope that he is, as I speak, turning his mind to the changes those noble Lords said the Bill required; if not, I fear we may need to change the Bill for him.