Dangers Of Law Clerks With Checklists

The discussion surrounding the dangers of automation typically centre on novel and futuristic application of Artificial Intelligence.  Even if we are not worried about Skynet terminating humans, we worry about application of data, privacy or robots making decisions.  But to the extent that these constitute problems that is not because of Artificial Intelligence, it is instead because of the problems of the relationship between technology and humans and this means that we should not fear for a future dystopia – we need to be concerned now!  That is, automation has encroached upon the law in many negative ways already and we need to fix it.  Advances in technology will not create some new problem, but it might accelerate existing problems that we already have. We can see these problems when law clerks use checklists.

Equity law is one of the “Priestley 11”, the compulsory subjects that all students must complete in order to become admitted as lawyers.  Therefore, at least theoretically every law graduate should understand what a trust is, the different types of trust and what a beneficial interest is.  Apparently not, it seems.  Although possibly some knowledge is crushed out of them by a corporate machine this apparent lack of knowledge is evidenced by the entirely misconceived questions that come from the legal department of institutions that review trust deeds. Often the review is done by a fresh-faced law graduate wielding a checklist of questions.

For example, it seems rather a regular occurrence to request a list of all of the beneficiaries/beneficial interests of a discretionary trust.  It must be that this question has appeared on some checklist, ostensibly for the purpose of ‘Knowing your customer”.  I am utterly at a loss as to what the correct answer might be.  For those who are faced with this question from the All High Law Clerks With A Checklist, who seemingly sit above the High Court on matters of trust law, the following are ­apparently not the correct answer:

• There is no list of beneficiaries required for a discretionary trust since McPhail v Doulton [1971] UKHL 1.
• There are no beneficiaries of a discretionary trust, merely “potential beneficiaries”.
• Potential beneficiaries do not have a beneficial interest in the trust property.
• There is no requirement for both a legal interest and a beneficial interest. That is to say, the beneficial interest is held by no-one.

Notwithstanding that these are all correct answers, there is functionally a “program” that exists that requires a certain input in order to proceed.

The only acceptable input is one that is wrong in fact and law, and the client is required to sign a statement attesting to the correctness of that incorrect fact.  And of course the responsibility is on the client to provide the correct answers to the questions and liability immures on them if they answer incorrectly or inaccurately.

Another interesting dictate of the All High Law Clerks With A Checklist is regarding bare trusts, especially in Limited Recourse Borrowing Agreements – which I would have thought would be much easier to understand given their simplicity.  It seems not.  Apparently bare trusts cannot be established with a notional settlor in the same manner as other trusts i.e. an unrelated party gives a nominal sum such as $10 to the trustee to establish the trust (this peculiar Australian method of settling is because of s102 of the Income Tax Assessment Act 1936 which could charge penalty tax rates of a trust established for the benefit of the children of the settlor).  Instead the All High Law Clerks With A  Checklist require that there is no property in the bare trust other than subject matter of the bare trust.  This has resulted in one of two egregious errors appearing extensively in bare trust precedents in Australia:

1. there being no property in the trust at its establishment. The trust therefore fails one of the 3 certainties required for validity: certainty of subject matter.  Therefore, the trust is not in existence; and
2. a valid trust comes into being after the acquisition of the subject matter of the trust. This of course leads to a separate dutiable transaction (i.e. double stamp duty) as per Farrar v Commissioner of Stamp Duties (1975) 5 ATR 364.

There is of course a relatively simple method of avoiding both of these problems and that is to pay the (future) trustee a nominal sum so to bind their conscience to hold the subject property on trust from the moment they receive it.  The trust therefore springs into existence with only the subject property, and because the trustee never obtains the beneficial interest in the property there is no second dutiable transaction.

One of the most perplexing dictates from the All High Law Clerks With A Checklist was for all the potential beneficiaries of a discretionary trust to go as guarantor for a loan taken out by the trust.  After explaining McPhail v Doulton and the basic nature of a discretionary trust, and hence why this was an impossible task (let alone unreasonable) I thought that I made some headway when the requirement was altered: a guarantee was required merely from the persons named in the discretionary trust around whom the class was based.

Unfortunately, the fact that those persons were long dead did not matter to the All High Law Clerks With A Checklist.

Nor did it matter that the controllers of the trust (who all had substantial assets) who were actually the primary default beneficiaries were guaranteeing the loan.  The signatures of the dead were required.

In an unusual effort at humanity the All High Law Clerks With A Checklist suggested that I could change the beneficiaries of the trust – remove the listed primary beneficiary and replace that name with their children’s name.  That this would totally change the class of potential beneficiaries under the trust and cause a resettlement (and Capital Gains Tax and Stamp Duty) fell upon deaf ears.  However, given that the programming of the All High Law Clerks With A Checklist required only the removal of the name of the person, I varied the trust deed so that the name of the deceased no longer appeared, but the trust and the class of beneficiaries was actually exactly the same.  This version (which had no substantive change) ticked the requisite box on the checklist and the dead were no longer required to sign.

Even though a human may be involved, if they relegate their decision making or thinking to some machine then the problems of automation will rear their head.  In each of the above examples the automated checklist should have been overridden by a human exercising judgment, lateral thinking, and contextual reasoning.  These are skills that humans are, and always will be, better at than machines.  The problems of automation are already upon us, they can be dealt with both now and into the future by understanding the proper respective roles of humans and technology in their symbiotic relationship.  Yes, it is true that a machine comprising a corporate system (such as a checklist) may exercise processing speed, reduce cost and create consistency and reliability.  However, a well designed system will have suitable safety valves so that control may be passed to the human.

I am reminded of Lord Denning’s description of humans and machines in Thornton v Shoe Lane Parking [1970] EWCA Civ 2:

“None of these cases has any application to a ticket which is issued by an automated machine.  The customer pays his money and gets a ticket.  He cannot refuse it.  He cannot get his money back.  He may protest to the machine, even swear at it.  But it will remain unmoved.”