To Deter or to Catch a Thief?
The conflicting goals of some transparency initiatives
There’s a scene in the Netflix romcom film “The Lovebirds” where the protagonists Leilani and Jibran, attempting to clear their names in a murder case, find themselves at the meeting of a secret society preparing to engage in an Eye’s Wide Shut-style orgy. As is common in secret societies, everyone is wearing a mask to ensure their anonymity. At some point, the emcee halts the proceedings and asks everyone to take off their mask. Both Leilani and Jibran oblige, but immediately realize that they are the only two to have done so. It turns out that the cardinal rule of the secret society is to never take off your mask and, by breaking it, they’ve revealed themselves as outsiders.
Before our protagonists can be harmed an alarm sounds and everyone wearing a mask rushes off. Moments later the police arrive and arrest Leilani and Jibran, who are the only two people remaining. The authorities have succeeded in catching someone - but anyone who knew any useful intelligence about the secret society kept their mask on and left out the back door.
The problem the police in The Lovebirds face is a common one for law enforcement: attempts to uncover information (gatecrashing the secret society’s orgy) changes the nature of the data you are gathering (everyone leaves). This is because those who have the most to lose from your information-gathering exercise are going to be the last ones to stick around while you conduct it.
This applies more broadly to efforts to fight financial crime and tax evasion. There’s been a big effort in recent years to enforce transparency in many different corners of the international economy. Some jurisdictions have begun gathering detailed information on foreigners’ bank deposits and have started transmitting that data to the relevant tax authorities. Others have started asking companies to reveal who their ultimate ‘beneficial’ owner is - a policy the United States is soon to implement.
These efforts generally have two goals:
Deterrence: Make those parts of the economy less hospitable to those who are trying to evade tax or earned their money illegally (and thus reduce the incentive to engage in these behaviors in the first place).
Information: Gather information that law enforcement/anti-money laundering agencies/tax authorities can use to follow up on and potentially prosecute cases.
But whenever the ‘bad guys’ are given a chance to close up shop before transparency policies kick in, one has to question the usefulness of the information that is gathered. For example, the introduction of the UK’s “People with Significant Control” (i.e. beneficial owner) register, managed by Companies House, arrived nearly three years after the UK government publicly committed to introducing it. Similarly, most countries that have started sharing taxpayer information with one another only do so months to years after the agreements are signed into law - giving tax evaders plenty of time to move their affairs to a more accommodating tax haven.
Of course, not all criminals and evaders are rational. Some will fail to effectively cover their tracks. Others will bet that the authorities won’t be competent enough to use newly-collected information effectively. The latter appears to be true at least part of the time - several investigations of the data filed with the UK’s Companies House have found significant gaps in the data quality, instances where companies have filed information that is very unlikely to be true - practically daring understaffed and under-resourced authorities to challenge them on it. This is the equivalent of showing the police a fake ID when they crash your party, rather than than quietly slipping out the back door.
But even if you believe some criminals stick around when transparency is introduced, surely the worst offenders got out long before. For example, the UK recently changed the law so a particular type of legal vehicle that had been tied to a lot of money laundering cases - something called a Scottish Limited Partnership - was required to reveal its beneficial owner information.
As a result, new SLP incorporations plummeted (see the graph below - for data reasons we can’t accurately see what happened to closures - only new incorporations). Looking at the graph, it’s easy to believe that bad actors stopped using SLPs after the law was changed. But if you believe that, you can’t also believe that the beneficial ownership information for new SLPs will be all that useful for chasing down bad actors.
However, there are instances where you can get the best of both worlds: where transparency is introduced as a surprise. Relative to convictions secured using publicly-available information,* data leaks frequently leads to consequences for criminal behavior. The Panama Papers scandal led to charges in the US for tax evasion, the sentencing of former Pakistani Prime Minister Nawaz Sharif for corruption and the reclamation of over one billion dollars in tax revenue by different governments around the world. Similarly, Isabel dos Santos - the billionaire daughter of Angola’s former president - had most of her assets frozen after a leaked cache of documents revealed the degree to which she abused her connections to government to amass her extreme level of wealth. If the companies that dos Santos had controlled had been given a year to reveal she was their ultimate owner, is there any doubt that they would have closed down, or just put down someone else’s name?
The more we can build in the element of surprise into new transparency efforts, the better. It creates a massive headache for financial institutions and governing bodies, but I think any new regime such as a beneficial ownership registry should be retroactively applied to any company that closes ahead of its introduction. Those companies should be immediately audited to obtain retrospective beneficial ownership information to prevent the worst offenders from escaping the new regime.
The aim of this post isn’t to argue that transparency efforts never result in actionable intelligence on money launderers and tax evaders. They obviously can: law enforcement frequently mention that they successfully connect information gathered from transparency regimes (e.g. transaction reports submitted to US Treasury from banks or beneficial ownership information submitted from areas with a high risk of laundering money through property) to ongoing cases. However, they key word there is ongoing - these are individuals and entities that were already under investigation.
But we have to recognize that this inherent trade-off between deterrence and the quality of information exists. It will help us better design future transparency efforts, and be more realistic about what we can expect out of them.
* That isn’t to say that sifting through public data isn’t important and doesn’t highlight abuses of existing regimes, reveal awkward arrangements or add evidence to existing cases. The OCCRP’s recent (excellent) work with Luxembourg’s beneficial ownership register has succeeded in all three of these things.