Contract analysis

The Devil’s Advocate™ Go Premium Sir! Sir! I've found an indemnity! In which the curmudgeonly old sod puts the world to rights. Index — Click ᐅ to expand: The Devil’s Advocate Comments? Questions? Suggestions? Requests? Insults? We’d love to 📧 hear from you. Sign up for our newsletter.

Luminaries, thought leaders and digital prophets will tell you that machines can now read and annotate contracts, such that yon poor legal eagles are no longer needed and will shortly be looking for work as pleasure droids or something. Quoth a such digital prophet:

Besides manipulating the physical world, machines are also increasingly encroaching on tasks that, until now, have required a human ability to think and reason. In the legal sphere for example J. P. Morgan has developed a system that reviews commercial loan agreements, it does in a few seconds it does what would have required, they estimate, about 360,000 hours of human lawyer time.

— Daniel Susskind, A World Without Work

Sounds fantastic, doesn’t it? Preternaturally intelligent silicon minds scanning and processing gigabytes of diverse text in an instant and analysing it for all material quirks and issues, like Zen from Blake’s Seven.

Data extraction tools

But let’s just stop and consider what is actually going on here. J.P. Morgan reviewing, presumably tens and probably hundreds of thousands of commercial loans (I am allowing JPM’s legal eagles, for whom I have great respect, ten hours to review each loan.)

These loans will be rendered in the hideous, overwrought prose of the banking lawyer, to be sure, but — even allowing for template variations and evolution (by which I mean periodic Cambrian explosions of flannelry — to which not even the House of Morgan is immune), the contracts will all be basically the same. There will be a predictable, schedule of customer details and economic variables — dates, amounts, currencies, rates, and optional elections — but the legal terms will be homogeneous.

So, “reviewing one hundred thousand commercial loan agreements” really means processing one hundred thousand tables of variables. This, doubtless, would take an aeon if assigned to a battalion of legal eagles, and would in the process drive each of the poor saps up the wall, but to be clear it would require absolutely no judgement, let alone legal acumen. It is a matter of transcription.

Assigning a lawyer to this task would, in any era, have been an act of cruel & unusual punishment, not to mention economic folly. It is not news that difference engines can process information better than meatware. JP Morgan’s sensible use of information processing power is hardly a horseman on the lawyer’s ridge.

Contract review tools

More potentially transformative are emerging “artificially intelligent” contract review tools. These promise to take a contract draft received from a counterparty, and analyse it against a pre-defined playbook, together with other examples which the machine has already learned from, mark it up, and return it for completion of the negotiation. A common subject for these contract review tools is confidentiality agreements.

But even here the technology disappoints and, in predictable ways, makes an existing problem even worse than it otherwise might be. Institutions are decommissioning review systems the same way they are repatriating once offshored operations functions. It is worth investigating why.

The problem statement

Start with the problem the contract review tool is designed to solve: Confidentiality agreements are seen as generally low-importance, low-risk agreements — assuming they are properly negotiated — that one must get through as quickly as possible in order to get on with more fruitful commercial negotiations. They tend to be “I’ll show you mind if you show me yours” kind of an affair. But — especially in this data obsessed world — they are buried risks if you don’t watch them carefully. The JC has a whole confi anatomy you can peruse if you would like to know more.

So you need your legal eagles to be en guard to stop the stupid getting in.

Now: confidentiality agreements come in all different shapes and sizes. They needn’t be longer than a few paragraphs, but our American friends are given to presenting 15-page bunker busters, which in the main amount to much the same thing as a concise one, but you never know and you must watch them like a hawk. There are a few points you need to check: the definition of confidential information, exclusions from the general definition, rights to disclose information that is confidential, the term, scope of remedies for breach. It is all, in concept, standard stuff. But it is a faff — depending on how excruciating is the writing style of the author, it might take up to an hour to review, correct mark up and send back. And it isn’t exactly glamorous.

So the problem: it’s slow, it takes a bit of time, the review parameters are fairly complicated (but not complex) — there is a limited risk of missing something — and ones inhouse legal eagle has invariably got better things she could be doing. This is, in person hours, effectively costing the firm money.

The contract review tool as a solution.

The contract review tool promises to perform this first basic check by reference to a pre-defined playbook of confidentiality policies, rather like a triage unit at a military hospital. You give it a once over and it goes out the door. Brilliant.

But it isn’t as simple as that

But it turns out parsing text lovingly confected in the brow of some unknown legal eagle isn’t as easy as all that. Just boring syntactical things like handling plurals and irregular verbs is extremely hard to code. And while in time the machine will get better at that, the universe of possible ways of articulating an idea is infinite and, while most legal eagles suffer from a form of locked-in syndrome which inhibits access to large parts of their imaginations when it comes to exercising their talent for textual complexity and saying things which, for the avoidance of doubt, are not in doubt, it is an information superhighway. Parsing the fevered prose of a human lawyer will always need some kind of a human sense check. So, the contract review tools build in exactly that kind of sense-check function. They hire paralegals, in low-cost jurisdictions, to check the output before sending it back. This has two unfortunate consequences:

• It is slow: Firstly, it slows down the output: Instead of getting your markup instantaneously, you get it three quarters of an hour or more later. This is more than enough time for the modern eagle to have been comprehensively distracted by something else, and in any case it is longer than it would have taken to review and fix the contract in the first place.
• It is more expensive: Secondly, it adds to the cost. Now, to be sure, reg tech providers are past masters at rent-seeking anyway, but here they have an actual out-of-pocket cost. Thus, the contract review tool will carry a heavy charge per document review. Better ones charge less than a hundred bucks. Some charge as much as three hundred, per review. Suddenly that cost proposition looks a bit squiffy: Your starting assumption is you are saving an hour of your internal legal’s time, which you unitise at say $250. But that is a nominal cost. Internal legal eagles are a fixed cost. They sit there, accruing expense, whether you use them or not. Most work hard, of course — it is a rare legalo eagle that punches out at 5pm on the smacker any day, let alone every day. That extra confi will get looked at during the day, at some point, anyway. Actually, reviewing confis can be a pleasant distraction: a nice par three recharger after a hard morning slogging up a series of long fairways on regulatory change stakeholder calls. (Management team: why don’t you try to get rid of all the stakeholder management calls?) So unless you can prove that, with your contract review tool you can actually let some of your lawyers go, it is not saving any money. It is costing money — more than it would have cost your internal team just knock off the confi in the first place.

See also

• Why is reg tech so disappointing?
• A World Without Work