What makes a good company register? Part 1: The public purpose

This is the first in a series of posts on company registers – why they exist, what they should do, and how they should do it, based on OpenCorporates’ interactions with over 70 of them.

Over the past six months, we’ve been asked again and again: What does a good company register look like – what should it do, what are some good examples, do we even need one? Most recently we were asked to feed into UK Companies House’s strategic review, asking precisely these questions.

So in this series of posts, we’re going to address these questions, going back to first principles and bringing them right up to date too, for the globalised, highly complex networked world we live in today.

So, let’s start at the beginning.  Why have a company register in the first place? It’s not, contrary to what some may have you believe, to generate revenue (we’ll come onto that later), but because companies are artificial creations, created out of thin air, and given a distinct legal personality separate from that of their owners or managers.

Why does it need a legal personality? Well, if it didn’t it’s just a collection of people working together (in some countries this is a de facto partnership), personally liable for what they do, and jointly responsible for assets and liabilities. A company allows the creation of an imaginary legal construct, owned and controlled by others, and able to enter into contracts, raise money, have assets and liabilities, and employ people. [Like fiat money, it’s one of those things we take for granted but is actually rather more conceptual than most people think.]

Until the middle of the 19th century these were very rare things indeed – in Britain, for example, until 1844 they had to be formed by an Act of Parliament or a Royal Charter. But the arrival of the railways, and the need to raise money for the people building them, helped create the push for the joint-stock enterprise, and not too long afterwards, the limited-liability joint-stock enterprise.

If you think creating a legal personality is strange, the idea of limited liability is even more so, as what it means is that when the liabilities (debts) of a entity exceed its assets (the money the shareholders have put in and anything that’s been accrued since then), then the people that bear the shortfall are not the shareholders or the managers (who benefited from the profits), but the suppliers (who don’t get paid), the customers (who don’t get the goods they’ve paid for), the employees (who don’t get wages), and wider society (lost tax, bailed-out pension funds, benefits, cleanup of polluted sites, etc).

So strange is this, that when it was suggested many were horrified, as it not only seems counter-intuitive, but fundamentally unfair, and also offer huge potential for fraud. Gilbert and Sullivan even wrote an opera about it.

Yet the belief was, that by allowing these to exist, there would be more innovation, more investment – in short society would benefit. But that last point is a critical one – the justification behind them is not for the benefit of companies, or their owners, but the benefit of wider society.

Critical to this is the ability to take an informed decision about whether you want to do business with the company, work for it, or, in the case of the state, ensure that it does not engage in fraud or other criminal activities. As Robert Lowe, the Vice-President of the Board of Trade, said when introducing the 1856 Companies Act, it was essential to give “the greatest publicity to the affairs of such companies, that everyone may know on what grounds he is dealing”. Hence the need for a company register – a public repository for that information.

When companies were by-and-large simple, straightforward things that were the legal embodiment of a bricks-and-mortar establishment and records were kept on paper this was straightforward. The state needed a record of what artificial legal entities it had created, and the public had a right, and a need to access this. In some jurisdictions this creation was performed by local courts; in others the power was delegated to a business register, a ministry of some sort, or, in the case of most US states, the state’s Secretary of State.

Either way, there was usually a central register for this information (although shockingly, and tellingly, Greece still does not have one, still less an open one). The difficulty of understanding what a company was reduced by the simplicity of companies (although there were exceptions) and paper-based records meant that there was largely equality of access for all, and of course increased the costs of companies having an overly complex structure.

Today we’re in a very different place, where it’s common for large ‘corporations’ to be made up of thousands of different legal entities, in scores of different jurisdictions, as our forthcoming hierarchy work, kindly funded by the Sloan Foundation, will show. It’s also common for companies to be set up entirely electronically, within the space of minutes, and for just a few pounds. And it’s common for corporate legal entities to be used to facilitate corruption, organised crime, money laundering, stolen assets and fraud.

In this context, it’s important to go back to the reasons why company registers exist: to be a public record of the creation and existence by the state of artificial entities that should ultimately be for the benefit of the state’s citizens. This is not possible unless “everyone may know on what grounds he is dealing”, and in the 21st century when our lives are not just governed by data, but are data, this means getting the information as free and open data.

Anything else has two negative consequences. First it effectively restricts effective use of the information to those with deep pockets (such as the traditional business information providers), who create rent-seeking business models, restricting innovation, enabling market inefficiencies and providing perverse incentives for complexity and opacity.

Second, in tandem with this it restricts the purpose to which the information can be put. It restricts it to those who can build a business model around it (and the financial crisis is a good example of a situation enabled by data that was used only by those who had the same narrow interest in that data), excluding journalists, civil society, employees, even government itself (which is very data-poor in this area). But it also restricts the ability to do cool things with the data, particularly combining it with other data to do data journalism, analysis and mashups. This, we would argue, is very much not in the interests of society.

Of course some governments can and do charge for access to the data – a shockingly large number, as our reports on access to company data in Open Government Partnership countries and the European Union graphically demonstrated. Governments could charge for a lot of things – in London the fire service was originally a private service run by insurance companies before it was realised that a burning house is likely to damage more than just itself. In fact one of the reasons for governments to exist is to solve collective-action problems, and to provide a base underpinning to society (the rule of law, for example). This is why we don’t charge victims of crime for their cases to be investigated, why we don’t charge for voting, and why, for example, US Federal information is in the public domain.

So why do company registers charge for the information? In some cases, this is due to historical quirks, in some it’s with the misguided idea that just by bringing money in the door they are doing A Good Thing, forgetting that in business it’s not just the top line that matters, it’s the costs too, and these are considerable to society by taking such an approach (which is why an increasing number of governments are starting to realise this by opening up their company data).

In others, it’s something rather more calculating and troublesome. Underlying this is the changing nature of what we mean by a jurisdiction in a highly connected world, given that governments are, in theory at least, bounded by jurisdictions, but global corporations are essentially supra-national, being everywhere and nowhere at once, as the debate about whether BP is a British or a US company (BP plc is domiciled in England, but has more US shareholders, employees and operations in the US).

This creates the opportunity the opportunity for companies to arbitrage jurisdictions and for states to compete with each other to attract a slice of the pie. Now, to a degree there’s nothing wrong with this. Countries compete all the time, investing in education, infrastructure, even immigration to create a beneficial, prosperous and attractive society.

However, there’s a second type of activity, which states sometimes engage in, where they deliberately perform acts which are knowingly and directly at the expense of other states. We’re no experts in this area of geo-politics, but these include things such as industrial espionage, state-backed hacking, even currency manipulation, and these are generally regarded as hostile acts.

Similarly, when a state sets up its corporate law to enrich itself knowing that it will benefit only by impoverishing other states, that too could be constructed as a hostile act. Now this isn’t an article about tax avoidance, transfer pricing, nor about money-laundering and global fraud, so we’re not going to dwell on these aspects.

But there’s a critical point here that relates to company registers. That is that all this behaviour can only be maintained in the shadows, as it is largely indefensible in the bright light of day, and the by-product of this is that such jurisdictions are beloved by criminals, fraudsters, money launderers and crooked politicians. The stories generated by the ICIJ investigation into the British Virgin Islands provide one example, but here’s an even simpler one, taken from forum dedicated to dedicated to “Offshore Company Formation services and Offshore Banking”.

The discussion is about the Seychelles offshore company register, which unlike the the onshore register(which is on the web and searchable) is pretty much entirely opaque, and has very few requirements, one of which, however, is that the companies registered there cannot pursue business within the territory of the Seychelles. They also allow bearer shares (and so the owner can change as easily as moving these around), and have virtually no filing requirements. In many respects, therefore, they are invisible, can’t be tracked, can’t be identified. Some – anti-corruption investigators, those fighting money-laundering, organised crime, fraud etc – may view this with horror, as a charter for any activity, legal and illegal, that you can think of. Others as a green light to do whatever they want:

What is happening here is that company registers are used as tools to sustain country business models that skim off a tiny slice of the money flows that can be diverted via them, via fees and a small – relative to the amount of money, but large for a country its size – industry of corporate lawyers and service agents. The fact that it does damage to others is not just irrelevant, but a requirement of the model, and well understood, as the prohibition of doing business in the home territory shows.

In other words we’re talking about the business model of pirates, operating offshore, but tacitly and not so tacitly supported by the country as long as they don’t attack the home country, and spend a bit of their loot locally.

Finally, this is a game that is unsustainable long-term, as it leads to an inevitable race to the bottom, and in that game there are only a very few winners. It has risks too, as  parasitical behaviour is a difficult balance to strike – risking the parasite either being attacked by the host, or killing it. Already we are seeing disquiet in Europe about the position of Luxembourg, which has gone from the poorest country in the community to the richest, largely through such a business model. And we would wonder how long before Delaware either comes under sustained attack from the other states, or realise that it’s not in their long-term best interests to be associated with the criminal behaviour its opacity enables.

So which company registers perform well, and which perform badly, from a public-purpose perspective? Well, the stand out is the New Zealand company register, which makes everything freely available, without any significant exemptions, and in our dealings with them have been clear in their position as a public register with a public purpose. That’s not to say there are no problems with New Zealand companies, but we suspect that one of the reasons they come to light is the ease of access.

The UK has made some positive moves, and is currently taking the lead in Europe, which, with the exception of Denmark and Norway remains in the dark ages on this (especially Spain, Italy, Germany and Greece). However it has a way to go to live up to David Cameron’s commitment to “shining a light on company ownership, land ownership and where money flows from and to”, particularly releasing the data Companies House holds on shareholders and company financials. It also has not yet taken that critical step of standing back and looking at the company register from a public purpose perspective (and of course its support for opacity jurisdictions such as the British Virgin Islands is an internal contradiction it has yet to come to terms with).

In the US, none of the states appears to be taking a public lead on this, not even California or New York which might be expected to. In fact, one of the best registries is Alaska, with Texas and Illinois, Nevada and Delaware the worst. 

Latin America, Asia and the developing world is very mixed, but a special shout out should go to Samoa, which has one of the best registers we’ve seen anywhere in the world, and one that puts many of the major countries to shame.