1901 Census

– in the House of Commons at 10:15 pm on 4th February 2002.

Alert me about debates like this

Motion made, and Question proposed, That this House do now adjourn.—[Mr. Ainger.]

Photo of Edward Davey Edward Davey Shadow Chief Secretary to the Treasury, Shadow Minister (Olympics and London), Liberal Democrat Spokesperson (Olympics and London), Liberal Democrat Spokesperson (Chief Secretary to the Treasury) 10:28 pm, 4th February 2002

Tonight, I am raising an issue that I can genuinely say is of interest to millions of people in the United Kingdom and across the world. About 30 million people sent a powerful message to the Government when, from 2 January this year, they tried—mostly in vain—to connect to the 1901 census website. Overwhelmed, the website closed on 6 January. I hope that the Minister will reassure me and those millions of people that the Government have heard their message and recognise not only that Ministers should ensure that a relaunched site works but that there is a huge opportunity for the Government on the web if they get this right.

For hon. Members who are not sure what I am talking about, let me briefly explain. The 1901 census was published on 2 January 2002. The Public Record Office met its statutory requirements to make that information available, and the public can now access the census using microfiche from the Public Record Office at Kew as well as from many local record offices and public libraries up and down the country.

This year, however, census publication was supposed to be different. As well as using the common microfiche format, the plan was to make the 1901 census available online. That was a fabulous concept. It would have enabled the whole world to search a database of 32 million names from Edwardian England, a transformation for historical research and a dream come true for the fast-growing number of professional and amateur genealogists.

As a newcomer to addictive family research, I should probably declare an interest. For nearly a year, I have dabbled in the search for my ancestors, and it was with some excitement on 2 January that I typed www.pro.gov.uk into my computer's search engine. Like millions of others, I was sadly disappointed. On behalf of those millions of frustrated people, I feel it right to raise the matter so that the Government may answer questions about the project.

I shall seek answers on three main points. First, how was the scheme's failure allowed to happen, and what went wrong? Secondly, how is it being put right, and is the Minister satisfied that it will be put right with no repetition of failure? Thirdly, what lessons have been learned? I believe that profound and far-reaching lessons emerge from what happened.

Before I approach the meat of those questions, I want to record my admiration for the Public Record Office. That may sound surprising given what I have just said, but I do not blame the PRO for the failure. If anyone should carry the can, it should be the contractor—QinetiQ—or the Lord Chancellor's Department and the wider private finance initiative programme. I have visited the PRO and seen the rest of its online work, and I know that it is a deeply impressive public service in which committed civil servants are dedicated to high-quality delivery for the public. One need only examine the PRO's annual reports and the extremely high customer satisfaction ratings that it achieves to realise that it is a public service that works. The PRO has done well when it has undertaken other information technology projects, such as its amazing online catalogue—PROCAT—or its access to archives initiatives. It has delivered them on time and on budget. The Minister may rest assured that I have no wish to have a go at the PRO. It is a cutting edge example of the public sector at its best.

My focus falls only on the 1901 census web-based project. The concept of putting the 1901 census online cannot be faulted. It was the right decision. The number of people who tried to visit the site shows that the public thought it the right decision. The problem lay in the execution of the project.

In 1998, it was clear to the PRO—presumably, therefore, to Ministers—that it could not pay for a project from its existing budget to put the 1901 census online. A decision was taken to opt for the PFI route. That initial decision raises a series of questions. Did the Lord Chancellor's Department ever consider funding the 1901 census as a one-off public sector project, led by the Public Record Office? What alternatives to the PFI were put to Ministers? Which Ministers took the decision? Was the e-envoy consulted?

One might understand why the PFI was considered suitable in 1998. It was presumably recognised that the website would be popular, and that money was to be made. In 1998, the dotcom bubble had not yet burst. The PRO went to tender in November 1998. I hope that the Minister will confirm that about 30 expressions of interest were received in response to the initial notice in the Official Journal of the European Communities. What I have not been able to discover is how many companies ended up on the shortlist. I understand that many companies withdrew when they realised the sheer enormity of the project—the transcription of 32 million separate 1901 census entries in a relatively short time, and the building of a website database system robust enough to withstand large demand. I am led to believe that the number of companies able to reach the final shortlist was extremely small. Will the Minister give the House that number? If she cannot do so now, will she write to me? The point is germane to the National Audit Office's future consideration of the project.

I have a sneaking suspicion that only one firm was on the shortlist: DERA—the Defence Evaluation and Research Agency—since transformed into the commercial, publicly owned company, QinetiQ. However, when the contract was awarded in October 1999, DERA was still an agency of the Ministry of Defence. My guess is that DERA was the default public sector option, as the private sector eventually declined to bid properly.

DERA still had to bid for a quasi-commercial contract to meet the tender standards and the so-called rigours of the Government's PFI. The Lord Chancellor's Department had to be seen to be participating in the PFI movement. DERA as QinetiQ is now truly commercial and may also soon be privatised. It may be bought by the US company Carlyle, whose chairman is former Prime Minister John Major according to The Sunday Telegraph and The Guardian. However, QinetiQ will never do more than the contract demands. It has no wider sense of the public good and no wider duty to the public good, although the census is public information and is thus almost by definition what economists mean by a public good.

What are the terms of QinetiQ's contract? Well, of course, we cannot be told—they are commercially sensitive. That was the frequent refrain in ministerial replies to my questions and those of Mr. Lidington, who is in his place this evening. We asked for details of the penalty clauses affecting QinetiQ—commercially sensitive. We asked for the final tender price—commercially sensitive. We asked for the contracts to be placed in the Library—commercially sensitive.

Will the Minister tell us whether we will ever learn the details of that contract? Will they ever become public and, if so, how long after the contract began? Will the 30-year rule—or perhaps even the 100-year rule—apply? Is commercial sensitivity always to be used as an excuse to prevent MPs from searching for the truth on behalf of the public good? Some of us think that commercial sensitivity is sometimes a pseudonym for "politically sensitive".

We know a few things, however. Since 1998, the taxpayer has spent £1.2 million on the 1901 census online project, but the bulk of the funding came from QinetiQ. It used independent finance, although no one would tell me how much. Perhaps the Minister can enlighten us.

We are told that QinetiQ is allowed by the contract to "cover their costs" and make "a reasonable return". We do not know how much the return might be. That is politically—sorry, commercially—sensitive. However, page four of the annual report of the Keeper of the Public Records notes that, after the contractor has made a reasonable return:

"Any additional revenue will be invested in digitising further census."

Will the Minister tell the House whether the digitisation of further census is dependent on what is left over after QinetiQ has made a reasonable return? If so, on current form, it might never happen, because we learn from the Minister's answers that the cost of the additional work needed to relaunch the site will be met by QinetiQ. At one level, that is reassuring, but there are two concerns.

If QinetiQ has to invest more to put the problem right, will it take the company even longer to make a "reasonable return"? Will it thus take even longer before QinetiQ can fund the digitisation of another census, such as that for 1891? One would need to see the detail of the contract to answer that question—but that is of course commercially sensitive.

Photo of David Lidington David Lidington Shadow Financial Secretary

Does the hon. Gentleman recall that the question of access to the accounts of publicly owned companies by the Comptroller and Auditor General and by the Public Accounts Committee was explored in some detail by the Sharman review set up by the Government? We have been waiting for more than a year for the Government's reply to that review. Does the hon. Gentleman agree that the census project would offer a good test case of the Government's readiness to allow the access by the Comptroller and Auditor General and the PAC that Lord Sharman recommended?

Photo of Edward Davey Edward Davey Shadow Chief Secretary to the Treasury, Shadow Minister (Olympics and London), Liberal Democrat Spokesperson (Olympics and London), Liberal Democrat Spokesperson (Chief Secretary to the Treasury)

The hon. Gentleman makes a good point. Lord Sharman carried out an important public audit review and I hope that the Government will meet its recommendations. The project would indeed make a good test case.

I want to return to the amount of investment. If the investment needed to put the website back online with a guarantee of good service was very large, QinetiQ may have been willing to make that investment, but no one may have been able to force it do so under the contract. That is a real concern, which takes me back to the inherent problem that lies behind this tragic episode—guessing the public demand for that new public service. As the apology on the PRO's website makes clear, the website was designed for 1.2 million visitors a day. On the launch, however, it received 1.2 million visitors an hour.

So we need to know how the projected demand figures were arrived at, what extra capacity will be built in for relaunch and what are the site's estimated settled or long-run capacity needs. Those were, and are, genuinely difficult questions, mostly because there has been no comparable site before. I am told that the PRO and QinetiQ looked at other sites, such as the American site on the passenger lists of ships arriving from Europe—the Ellis island site—and at the Mormon genealogical sites and that they used the figures from those sites, and added some, but that still was not enough.

Should they have built more capacity? We are also told that many genealogists were worried that site capacity was not sufficient and that they told the PRO that in October 2001. Yet the PRO is locked into a Government-required PFI contract, so it has little room for manoeuvre. If QinetiQ is not prepared to pay for the extra investment, it will not happen; it did not before the launch, and, worryingly, the extra investment may well not be there for the relaunch. Is the Minister aware that no new extra capacity is planned for the relaunch? Does she think that wise?

We have read about the increased bandwidth, the firewalls, the divert sites. I have heard that cookies might be developed to knock people off the site if they stay on for too long, blocking others. But what about extra capacity? The aim seems to be to make the site more robust for the 1.2 million estimated visitors to ensure that they have a good experience when they finally get on, but that may well mean long waits and long queues because there is no intention to increase the capacity.

Does not the Minister realise that the site's capacity will not increase unless and until she and the Department intervene to make that happen? The PRO simply does not have that power over its contractor. I urge the Minister to think long and hard about that. The House will hold her and her colleagues to account if the relaunch does not work. People want to know when the site will be up and working. Can she give the House and the millions of people interested in the site in the United Kingdom and around the world the categoric assurance that the site will work when it is eventually relaunched? When does she expect it to be relaunched?

In the short time that remains to me, I want to explain to the Minister why I think the project and its success deserve her attention, and, with due respect to her, the attention of Ministers at the very highest levels. The project deserves significant political attention because the site was so popular. When so many millions of people overwhelmingly show that they want information on their own kith and kin that the Government hold, surely the Government should sit up and take notice.

More people visited the site than watched "Big Brother", than voted on "Pop Idol" or than voted in the last election. The focus groups may not have told those at No. 10 that, but the Minister should do so. Here is a chance to put family values back into the internet, rather than the smut and pornography that has dominated it. Pornography may speak to a human desire, but the desire to contact our family—to reach out to our past and our own genes—is also a powerful human desire, and the Government can back that desire by backing the 1901 census website. Just as the website "Friends Reunited" has been so successful, appealing to the human desire to find long-lost friends, so the census appeals to another positive human instinct.

Let me put this huge opportunity another way. Rupert Murdoch realised that he needed sport to sell satellite television. He paid huge sums to win exclusive rights to screen test cricket and football, and it worked. With the public sector's monopoly over the census data, the Government could use the census website to build e-government. Let us compare the experience with that site to the hundreds of other Government websites. Yes, people find those websites useful, in their tens, in their hundreds and in their thousands, but which Government website has had millions of visitors in an hour? If we harness that public excitement and the curiosity that people have about their own families, we might transform the prospects and popularity of e-government. People who would otherwise never think of surfing the net might just try it out.

For example, I have tried several times to explain the internet to my grandmother, but she smiles at me and says, "Yes, dear. But it's so boring." However, when I talk to her about her parents, grandparents and our shared family history, she comes alive. I am not suggesting that I will ever get her surfing the net, but the 1901 census would certainly make it interesting for her. Therefore, for relatively tiny amounts of cash—especially in the context of the hundreds of other e-projects around Whitehall—the Government might just attract the interest of millions more to the web and eventually even to other e-government sites.

The 1901 census site has the proven potential to move online government on to a new, higher level—and cheaply—but only if the Government move fast and back the Public Record Office before the site's relaunch.

Photo of Rosie Winterton Rosie Winterton Parliamentary Secretary (Lord Chancellor's Department) 10:45 pm, 4th February 2002

I congratulate Mr. Davey on securing this debate on the 1901 census website. The site has certainly generated considerable interest over the past few weeks and I now realise the personal interest that he takes in the subject, about which he spoke passionately.

I am glad that the hon. Gentleman took the opportunity to praise the staff of the Public Record Office. They certainly deserve our praise, and I too would like to praise them for putting together this project and for doing so in the way that they did. The project is a cornerstone of the PRO's strategy to use information technology to make the historical records in our national archive more accessible to the public.

As the hon. Gentleman said, the Keeper of the Public Records has a statutory duty to make all census returns available after 100 years. That is her statutory duty, and in the past it was done solely by providing a service on microfilm. However, in planning for the release of the 1901 census, the PRO thought beyond that and wanted to come up with a new and exciting initiative. It wanted to provide an additional online service as it recognised that this was a time of rapidly expanding internet access. It also recognised that that presented a tremendous challenge and an opportunity. Making the census available in that way offers users, for the first time, the prospect of accessing information and images about the more than 32 million people who lived in England and Wales in 1901.

The hon. Gentleman asked a number of questions. I shall try to answer as many as I can but if I do not answer them all, I hope that he will feel free to write to me. I shall certainly respond as best I can, bearing in mind that issues of commercial confidentiality are involved. Some of the questions that he raised are covered by that.

Let us start from the basis of the hon. Gentleman's argument, which was about how the project was funded in the first instance. It was clear that the creation of 1.5 million digital record images and complex IT systems could not be funded out of the PRO's resources. Its annual budget is about £25 million, and there is no doubt that the project would have required upfront several million pounds of additional funding. He mentioned many of the PRO's other commitments and, because of them, it was clear that partnership funding would be required if we were to get this exciting project off the ground. The PRO therefore rightly sought to establish a public-private partnership, the procedures for which it followed to the letter at all times.

I note the hon. Gentleman's views on the decision to use the process, but I think that the right decision was made. It is a good example of how the public and private sectors can work together to bring service benefits to the citizen at a lower cost to the public purse and at a reasonable cost to the customer, whether accessing the website from the United Kingdom or overseas, because many people outside the country were expected to use it.

The PRO began the process in November 1998. About 30 organisations expressed interest in the contract, and a full tender exercise was undertaken. Four organisations were shortlisted to produce a full submission. One could not commit to making the 1901 census available online by January 2002 and another withdrew, which left two organisations to provide full submissions. That gives an idea of the scale of the project and the possible risk involved.

The contract was based on the standard Office of Government Commerce model for information technology projects. The contracts are outcome-based, and the commercial and development risks are transferred to the contractor. The PRO established a dedicated team to work closely with the contractor through a joint programme board to ensure that the deadlines were met. That included the successful delivery of a pilot project, which I saw, using the 1891 online returns for the county of Norfolk.

In addition, the PRO spent £1.2 million on essential work related to the 1901 census, such as quality assurance. I emphasise that it has not made over any public funds to QinetiQ. The development costs for the online service have been met entirely by QinetiQ using its own sources of finance; it has not received payments, management charges or other fees from the PRO. As the hon. Gentleman rightly said, QinetiQ recoups its investment costs through revenue on the charged services from the census website. Once the company has recouped its investment costs, it will pay a proportion of its income back to the PRO for investment in further online services, notably other censuses. In that sense, it is fulfilling the requirement, which the hon. Gentleman set out, that we should consider putting other censuses online. On the extra money that has been invested to improve the site, it is difficult to give the exact timing for the recouping of the money and so on, but we hope that we will put the 1891 and the 1881 censuses online in about a year.

An advisory panel comprising amateur and professional family historians as well as staff from QinetiQ was set up in January 2000. It met 11 times to give users a full say in the development of the online service. On estimated demand for the service, it is true that members of the advisory panel highlighted the problems that family history websites in the United States experienced at initial launch when attempted use was much heavier than anticipated. At the same time, the panel carefully considered the example of other US websites. It is always difficult to make precise comparisons because some sites provide different information to different audiences. However, the important lesson that was drawn from the US experience—this is at the very heart of many of the hon. Gentleman's questions—is that, after the initial surge in interest, demand for the service markedly declined.

I am sure that the hon. Gentleman would accept that it would not have made business sense, or any sense, to construct a service that was capable of meeting all demand experienced daily between 2 and 6 January because, in the long run, all the evidence is that demand falls off. That would therefore have been a waste of resources and unnecessary expense.

There were initial problems with the launch, but they largely arose because the service's popularity outstripped even the generous estimates of likely use. Up to 1.2 million users per day were planned for, which is a massive site by general internet standards. By way of example in the United States, the Genealogical Society of Utah's website, which is probably the biggest such site and, incidentally, had the same kind of problems to start with and crashed at its launch, now has 89,000 users a day.

We know that nearly 30 million users a day attempted to access the 1901 census between 2 and 6 January. We are working to put right some of the problems that have arisen. I assure the hon. Gentleman of that. Officials from the Lord Chancellor's Department are working with officials from the Public Record Office and staff of the company to ensure that the website is up and ready as soon as possible.

I am aware of the time, so I hope that I have covered a number of the issues raised by the hon. Members for Kingston and Surbiton and for Aylesbury (Mr. Lidington). I know that the hon. Member for Kingston and Surbiton has shown an interest in the work of the Public Record Office recently, and I know that he visited it last week. His hon. Friend Dr. Tonge has also shown a keen interest in and been very supportive of this ambitious and innovative project.

Of course I understand the frustration of those people—including the hon. Gentleman, now that I know of his intense personal interest—who were looking forward to finding out about their families and roots on the census site, but I firmly believe that it was right for the PRO to undertake the project. It will provide millions of people with the opportunity to access public records and provide a revenue stream to fund other online services. We would never have been able to do that without a partnership between the public and the private sectors.

I am very aware of the hon. Gentleman's concerns, and assure him that we are doing everything we possibly can to ensure that the census site becomes the success that it deserves to be and provides enjoyment and interest for the many millions of people who wish to use it.

Question put and agreed to.

Adjourned accordingly at two minutes to Eleven o'clock.