Tuesday 4 June 2013

He's all heart, Shakespeare

Forecasting future benefits is also hard to predict

Stephan Shakespeare, writing in An Independent Review of Public Sector Information, devotes one section to the question who owns public sector information.

In the opening paragraph of the section entitled Ownership (pp.28-33) Shakespeare says: "I think the time is now right to reflect on how the current models of ownership apply in the current context" (p.28).

We know from an earlier post the current result of his current reflections – please see Shakespeare's take on property. He wants to give all public sector information (PSI) to "businesses, especially SMEs". For free. Without charge.

What we're looking for in this section of his review is the reason for Shakespeare's recommended largesse. We're dealing with ownership here. Property law. And we're entitled to some more or less scholarly argument.

----- o O o -----

Shakespeare begins by discussing the data managed by Companies House, the Land Registry, the Met Office and Ordnance Survey. These four together are apparently known as the "Public Data Group" or "PDG" for short.

Take Companies House as an example.

As everyone knows but Shakespeare doesn't mention, at the moment:
  • You pay to set up a new company or you buy an existing company off the shelf.
  • You pay to run the company.
  • You pay to prepare the accounts, following the rules laid down by Companies House, HMRC, the Companies Act and any number of national and international accountancy standards organisations including the OECD.
  • You may pay to have the accounts audited. The auditors are highly trained, regulated by several accountants' institutes and they pay for hugely expensive professional indemnity cover.
  • Or, if you are conversant with section 477 of the Companies Act 2006 and it applies, you pay to hold a general meeting of the members of the company and you try to agree with them to dispense with the audit.
  • You learn how to use iXBRL and you submit your accounts to HMRC and pay your corporation tax, if any, being wary of the GAAR, and you submit them to Companies House and you pay the Companies House fee for making an annual return.
  • You can be fined for making a late return, you can be forced to re-submit your accounts if you make a mistake, you can be fined or banned from being a director if you misbehave and your auditors can be fined or struck off if they misbehave.
By the time your company accounts poke their head through the mist and into the public gaze – anyone can see them if they only pay Companies House £1 for the privilege – they contain a lot of personal information and a lot of people have made a lot of effort and taken a lot of risks and paid a lot of money to get them there. There are a lot of barriers to entry.

Once they are public sector information on the Companies House website, who do the accounts belong to?

Companies House? HMRC? The company? The members/shareholders? The auditors? The accounting institutes? The professional indemnity insurers?

Shakespeare doesn't tell us. We learn nothing from him about ownership.

He makes vague allusions to them belonging to "citizens".

What he is clear about is that this data should belong to "businesses, especially SMEs" who know nothing about the company, have taken no risk, have made no effort and will make no payment for the data even if they could afford to because that would constitute a barrier to entry.

Why does he want to give them the data?

Because, according to Shakespeare, some good may come of it. What good?

Suppose that the PDG charged nothing for public sector information (PSI), Shakespeare says. Then:
As government would no longer need to purchase the PSI itself, the direct loss to the Exchequer on an annual basis is in the order of £143 million ... It seems a straightforward decision to invest £143m to make Trading Fund data widely available is a relatively small price to pay to leverage wider economic benefits far exceeding this by orders of magnitude. (p.30)
He wants to deprive the Exchequer of £143 million p.a. and expropriate the citizens' data in order to "leverage wider economic benefits far exceeding this by orders of magnitude". What "wider economic benefits"? No idea. How many "orders of magnitude"? No idea. When? No idea.

And yet he calls it a "straightforward decision". It's not a decision at all, is it. It's a straightforward hunch. A mere guess. An unsupported hypothesis. Feckless optimism.
The reforms that I have suggested should not result in an unjustifiably high cost to Government but putting a price on that is for Government to do. (p.30)
That may be how they do things at YouGov, Shakespeare's company.

Someone proposes an investment, Shakespeare asks the finance director if YouGov can afford it, the finance director says, "yes, we should be able to" and that's considered an adequate answer. Governments can't be quite so cavalier. They're dealing with public money.

And what's all this about "putting a price on that is for Government to do"? Isn't it for Shakespeare to do? Isn't that why he's been asked to perform his review? If he can't "put a price on that", how can he make his case? He can't.

He even tells us why he can't do his review job:
Forecasting future benefits is also hard to predict. How businesses and individuals might use datasets in the future to generate new products and services and by implication impact economic growth, is equally unknown. (p.30)
It's too difficult to do his job of forecasting. The uses to which "businesses, especially SMEs" might put public sector information are necessarily unknown. What "new products and services"? No idea. How will they "impact economic growth"? No idea.

But in that case, what justification is there for Shakespeare stating that the impact will exceed £143 million "by orders of magnitude"? None. 143 million times, none.

Shakespeare wants to change the "funding model" of the Public Data Group.

The data revolution is moving rapidly, and faster than government structures are reacting to that change. (p.29)
What data revolution? He doesn't tell us. Apart from Shakespeare, who says that "government structures are reacting to that change" too slowly? Why is it the right reaction to abandon agreed custom and give public sector information away for free?

The questions go on and on and never once does Shakespeare have an answer.
My conclusion is that to quantify the costs and benefits precisely from outside Government is difficult due to the many complexities, however, I think there is sufficient evidence to support the theory that the benefits far outweigh the costs to releasing, firstly data from the Trading Funds and secondly, PSI across the public sector.
The Shakespeare review provides "sufficient evidence" because Shakespeare thinks it provides sufficient evidence. This is the world of make-believe, not the world of public administration.

You may or may not agree. Tell us what you think. Please complete the poll at the top right of this web page. (Poll now closed. Results here.)

----- o O o -----

Shakespeare defines "public sector information"/"open data"/"big data" on p.8. We know what he's talking about:
PSI covers the wide range of information that public sector bodies collect, produce, reproduce and disseminate in many areas of activity while accomplishing their public tasks.
On p.31 he starts talking about something quite different – private sector data:
Almost all of my review is focused on increasing the availability of PSI, but there are also opportunities from opening up private sector data.
Then he stops again:
It also opened up discussions on who actually owns the data but I won’t go into that further now.
"I won't go into that now"? What's going on? If he doesn't want to go into the matter, why does he raise it in the first place? If he wants to talk about "who actually owns the data", what better place than the section on ownership in his review?

The answer is that he's warning everyone that his plans do not stop at public sector information. He wants his National Data Strategy to include private sector information as well. Including your personal data. He says:
There have been real transformational benefits from initiatives such as Midata where consumers now have access to their own information collected on them by retailers and others. That is a huge step in really empowering consumers to take decisions based on data that they themselves have generated. I'm sure that Tesco didn't design their loyalty card scheme with open data in mind but this has been a truly groundbreaking step in access to private sector collected information.
We know that that's false.

Long before midata was thought of by the Department for Business Innovation and Skills (BIS) or, possibly, the Government Digital Service (GDS), banks gave us bank statements, telephone companies gave us itemised bills, retailers gave us invoices, etc ...

midata is a false prospectus.

Like Shakespeare and his PSI, midata promises great benefits without ever being able to explain what they are – it's the South Sea Bubble all over again.

And like Shakespeare and his PSI, the man in charge at midata is Professor Nigel Shadbolt, please see midata – a machine for turning personal data into open data.

The PSI people have an unfounded belief that they know how to grow the economy. And the midata people have a megalomaniac belief in addition that they know how to run your life better than you do.

You have been warned. By Shakespeare. The revolution is coming.

----- o O o -----

Innocent people get badly hurt in revolutions. It is conventional to apologise for that. Some contrition is normally expressed for all the eggs that had to be broken to make the omelette.

But not old Shakespeare:
Those currently deterred by charges would benefit from reforms and conversely, organisations who are at an advantage in using their own proprietary information for commercial advantage, might find their competitive advantage diluted if more PSI is released. But in dynamic markets this happens all the time and is a stimulus for innovation and so business should embrace the change. (p.31)
Embrace the change.

His message to the 40,000+ public servants who will be laid off if GDS ever manage to get digital-by-default up and running? Embrace the change.

British steel plants shut down? Embrace the change.

Coal mines shut down? Embrace the change.

Upper Clyde shipbuilding shut down? Embrace the change.

He's all heart, Shakespeare.


Updated 5.6.13 14:24
Rather pleasingly, the real Shakespeare – William, not Stephan – is now following DMossEsq on Twitter. He reminds us all of a relevant passage from As You Like It:
CELIA ... not a word?
ROSALIND Not one to throw at a dog.
CELIA No, thy words are too precious to be cast away upon
curs; throw some of them at me; come, lame me with reasons ...
Come, Stephan, lame the lot of us with some reasons to believe that it's worth giving PSI away for free.

Updated 18.12.15

It's 2½ years since the post above was written. What progress? How goes the revolution? What changes are there to embrace? Where is Professor Sir Nigel Shadbolt's explosion of innovation? Thanks to Stephan Shakespeare, is it Christmas for the UK economy every day?

Let's take a look at Companies House, one of the four members of the Public Data Group, please see above.

And let's take a look specifically at the Find company information facility on the Companies House website.

Companies House have decided to stop charging anyone to see the returns filed by companies. It should be quick and easy to make that change, shouldn't it. You just cut out the charging processes in the computer system. But that's not how Companies House have tackled the job.

No, too quick and easy, instead they have created a new test/beta website, on which you can see these returns for nothing, unlike their current/old/live website, where you have to pay £1 to see each return.

Why did they do that? Why did they choose the harder option? This way, they have to maintain two websites instead of one. The new website lags behind the old one, it's still trying to catch up with the facilities already available in the old one. What's the point?

The answer seems to be, on inspection, that the new website includes facilities to investigate people. You can investigate companies on the new website, just as you can on the old one, but now it's much easier to see, for example, what other companies Stephan Shakespeare is a director of:
Stephan SHAKESPEARE - Personal Appointments (free information from Companies House) page #1 of 5
What innovative ideas have resulted from this revolutionary progress? None, so far as is known. Benefit to the economy? Ditto. "It's early days", you may say. But if not now, then when do we see the benefits? Never?

The effect of Companies House introducing their new website is slightly intrusive. Their response is to remove the day of the month from Mr Shakespeare's birthday which is given in the example above as April 1957.

"Well done, Companies House", you may say, "how courteous and sensitive". Your admiration is misplaced. Take a look at the YouGov annual return to 31 July 2014, for example, which you can do quickly and for free, and there's his full birthday. And several other peoples'.

The new website is more intrusive but at least the data is free. Except that it's not. Obviously taxpayers are having to pay Companies House to create this new website as well as maintain the old one. But that's not all. Here's a page from the YouGov filing history on the new website:

YOUGOV PLC - Filing history (free information from Companies House)

Companies House want you to pay £3 for that resolution to remove pre-emption rights. And you have to ring them for the privilege. Whereas, oddly enough, you can get it for £1 without making a telephone call, digital by default, from the old website:

"Forecasting future benefits is also hard to predict", as Mr Shakespeare so rightly says.

Updated 22.1.16

Companies House understand that they have made a mistake by publishing so much personal information for free on their proposed new website. That's why Our register: advice on protecting your personal information appeared on their blog yesterday. Anyone reading it will realise that it doesn't correct the mistake.

The following comment has been submitted. Let's see if it is published and if it elicits any sense from Companies House:
David Moss
Your comment is awaiting moderation.
Suppressing the day of the date of birth of any specified officer of a company on the new appointments query is the flimsiest of fig leaves – the full birthday is readily available for free in the annual return.
How does that fig leaf protect anyone's personal information? It doesn't.
The home addresses of the officers of a company may be hidden in future, but earlier returns are still there on the web, often complete with home addresses.
There was some protection for personal information when people had to pay £1 for each document and when it took some time to download and read them. That protection has been weakened considerably now that the results are available instantly and for free.
The reason for this change seems to be that making everything open and removing any barriers to entry will promote innovation and expand the economy. There is no sign of any such innovation.
As a matter of interest, are you the Ian Gronland at 97 Andersons, Stanford-Le-Hope, Essex, United Kingdom, SS17 7JD or the one at 23 Highfield Gardens, Grays, Essex, RM16 2NT or the one at 9 Riverview Flats, London Road, Purfleet, Essex, RM19 1SL?
Link to this comment
Updated 25.1.16

The comment immediately above hasn't been published on the Companies House website and there has been no response.

You may think that it shouldn't be published. It includes the postal addresses of one, two or three Ian Gronlands. Those should not be published free for all to see without the owners' permission.

But that isn't an argument Companies House can use. They are the ones publishing the Ian Gronland addresses and the addresses of hundreds of thousands of other directors and company secretaries. If it's wrong to publish the DMossEsq comment, it's wrong pari passu to publish the free-for-all-to-see beta version of the Companies House website without Gronland permission.

You may alternatively think that the DMossEsq comment hasn't been published because, in the event, there has been no avalanche of innovation inspired by the new Companies House website.

If opening up all this personal information hasn't allowed the UK to "leverage wider economic benefits far exceeding [£143 million] by orders of magnitude" as promised by Shakespeare, please see above, then creating this new Companies House website is a waste of time and money.

Updated 27.1.16

The eccentrics at Companies House
consider that fines
don't deter inconsiderate parking

Companies House have now published the comment above and responded as follows:
Esme Turner
Companies House does not consider that information is more readily available as the result of the removal of the £1 fee to access it. We do not consider that a fee of £1 would deter anyone who wants to access a piece of information. The decision to redact day of birth information from our data systems and to alter the design of forms so that for filings made after 10 October 2015 is deemed to be a proportional response to the threat of identity fraud. Given that this information is already in the public domain, even the removal of it from Companies House systems would not prevent third parties from using it.
We have not published the final sentence of your comment. Companies House does not confirm the addresses of our staff and we will not publish blog posts that may contain this information.
Link to this commentReply
They don't answer all the questions raised and the answers they do provide raise more questions. A further comment has been submitted:
David Moss
Your comment is awaiting moderation.
Thank you for your 25.1.16 response.
Companies House consider that removing the fee doesn't make personal information more readily available and that paying the fee wouldn't deter "anyone" from accessing the personal information recorded by Companies House. This is an eccentric position to adopt. Most people accept that penalties deter inconsiderate parking, for example. Companies House may care to consider the tweet sent to them last June by a blogger/journalist who claims to have scored a success using the new website and says explicitly "if I'd had to pay £1/document, I wouldn't have found the info".
It's not just removal of the fee that increases "the threat of identity fraud" which you mention but also the speed and convenience with which personal information can be collected on the new website. Users no longer have to select documents laboriously, as we did on the old website, log in, enter our payment details, wait for the documents to be marshalled, read them, go back and start again, ... Those disincentives have been removed and, instead, bingo, there it all quickly is, all the personal information you could want and more.
Companies House clearly recognise that they have put people's personal information at risk. Why else publish your blog post, 'Our register: advice on protecting your personal information'? Three people have submitted comments on your blog post airing their qualms. Omitting the day of people's birthdays from some documents but not others is only a "proportional" response if Companies House is once again being eccentric.
The claim that personal information is available elsewhere does not justify Companies House's making it available so conveniently. Some people are unrealistic, for example. Should Companies House follow their example?
Why are Companies House seeking to change the status quo with the new website they're testing? You don't unfortunately answer that question in your response. Many people have advanced the hypothesis that making information available to everyone will inspire innovation and cause the UK economy to expand.
Here it is being tested by Companies House. The hypothesis, that is. Has it worked? Has it inspired a lot of innovation? Has it caused the economy to expand? If not yet, then when? So far, in this case at least, it looks as though the hypothesis is wrong. In which case, why proceed as though it is right? Is it time to abandon the new test Companies House website?
Did Companies House perform a privacy impact assessment before spending time and money on a new website? If so, what were the findings? And if not, in view of the concerns expressed by commentators and by Companies House themselves, isn't it about time to assess the privacy impact?
Link to this commentReply

Updated 6.3.16

Where there should be a coherent argument,
there's just a hole.

As noted above, Shakespeare wants to make all the data stored by Companies House, the Land Registry, the Met Office and Ordnance Survey freely available to anyone who wants it. The effect would be, he says, for innovation to be inspired and for the UK economy to expand by "orders of magnitude".

Also noted above, Shakespeare offers not a single cogent reason to believe him.

Companies House have nevertheless wasted their time and our money on developing a new website which makes it quick and free to gather reams of personal information about the directors and shareholders of limited companies.

They realise there's a problem, which is why they published Our register: advice on protecting your personal information. Their advice is utterly ineffectual, but Companies House will pursue the Shakespeare strategy anyway.


No explosion of innovation has been detected since Companies House started giving away millions of people's personal information for free. The economy has not noticeably expanded as a result.

Where there should be a coherent argument, there's just a hole.

Far away from Companies House, the Land Registry and the others, the Government Digital Service are promoting their new identity assurance scheme, GOV.UK Verify (RIP). That scheme relies on appointed "identity providers" verifying who we are and issuing us with an identity so that we can access public services on-line.

How do the "identity providers" do that? How do they satisfy themselves on-line that we are who we say we are?

One of them, Safran Morpho, tells us:
1.3 How does Morpho collect your personal data
Personal data that Morpho may check, include:
- Your Credit Record History
- Your Electoral Roll History
- Your financial court orders records (CCJ, IVA, DRO, Bankruptcy)
- Your record in the Land Registry ...
- Your Directors Register record

We might in certain circumstances verify if you are active on social networks.

Morpho may collect personal data about you because Morpho is required or authorised by law to collect it.
Does that fill the hole?

In the absence of any other, is that perhaps the explanation for Shakespeare's and Companies House's delinquency?

No comments:

Post a Comment