By the time walking away is the right answer, you have usually spent money, months and a quantity of public commitment - and all three argue in the same wrong direction.
12 min read · Level 3
The standard advice is to ignore sunk costs, and as a description of correct reasoning it is unimprovable. As advice it barely works, and it is worth being honest about why. People do not continue failing processes because they have misunderstood the arithmetic - explain it and they will agree immediately, and then proceed anyway. They continue because stopping requires saying out loud that fourteen weeks of their own judgement were wrong, in front of people whose opinion of that judgement matters to them, and no amount of correct reasoning about sunk costs addresses that, because it is not the cost being defended. If that diagnosis is right, then the useful interventions are social rather than analytical: writing the conditions down in advance so that stopping is executing a prior decision rather than admitting an error, telling somebody in week one what would make you stop, and being the person in your own circle who has visibly walked away before and was respected for it. The implication is slightly deflating for anyone who likes frameworks. The discipline is not a better model of the decision. It is an arrangement made in advance so that the decision does not have to be taken by the version of you who will be in the room.
Stopping is a separate skill from investigating, and it is the one the process is worst at supporting.
Everything about a live transaction pushes in one direction. Costs have been incurred and they feel like an investment in the outcome rather than what they are, which is money that is gone in either case. Other people know - a partner, a spouse, a committee, the seller, two sets of advisers - and a decision to proceed requires no explanation while a decision to stop requires several. Time has passed, which makes the alternatives feel stale even though nothing about them has changed. And the work itself creates familiarity, which is very difficult to distinguish from conviction.
Against all of that stands a single sentence written before any of it started: the price above which, or the finding after which, you would not proceed. A walk-away price is not a negotiating position. It is a message from the version of you who had spent nothing, was not tired, and had no audience.
Because the arithmetic is not close, and almost nobody does it. Suppose a process has consumed 38,000 of legal fees and 22,000 of accounting - 60,000 spent - on a contemplated investment of 750,000. The 60,000 is 8.0% of the investment, and it is gone whether you proceed or stop. The 750,000 is not gone. Continuing in order to avoid wasting 60,000 means putting 750,000 at risk to protect an amount 12.5 times smaller than itself, and stated that way nobody would do it. It is never stated that way.
And because the cost of stopping is almost entirely a cost to your self-image, while the cost of proceeding badly is a cost to your capital. Those are not comparable quantities, but they do not feel that way in the week of the decision, because one of them is happening to you in a room with other people in it and the other is an abstraction about a future you have not yet met.
This is educational information, not investment, securities or financial advice. Rules differ by country and change — get advice from a qualified professional in your jurisdiction before acting.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Sources: U.S. SEC — Investor.gov; FINRA — investor information; UK FCA — information for consumers
Fourteen weeks in. 38,000 of legal fees and 22,000 of accounting work, so 60,000 spent. The contemplated investment is 750,000. The walk-away note written in week one said: 'I do not proceed if the revenue is materially concentrated, or if the licence does not transfer cleanly.'
Both have now happened. The top three customers are 60.0% of revenue, and the licence covering 69.0% of it terminates at the licensor's option on a change of control.
Here is how the conversation goes instead. Somebody observes that 60% concentration is common in this industry, which is true. Somebody else notes that licensors usually consent, which is also true and is a statement about a population rather than about this licensor. A structure is proposed: a lower price, part of it deferred, contingent on the licence being renewed. The structure genuinely reduces the exposure, and it is offered by people who are being helpful rather than reckless. And the 60,000 sits in the room without being mentioned once, doing more work than anything that is said out loud.
Now the arithmetic that is never performed. Proceeding at the revised price still commits 750,000. Stopping costs 60,000 and returns 750,000 of capacity to you immediately. For proceeding to be right, the revised terms have to make the investment genuinely attractive on its own merits - not less bad than it was an hour ago, which is the standard that is actually being applied, because the comparison being made is with the previous version of the deal rather than with everything else you could do with 750,000.
That is the whole test, and it has a specific form. Ask whether you would enter this transaction, on these terms, with these findings, if it were presented to you today for the first time and you had spent nothing. If the answer is no, the 60,000 is not a reason to proceed. It is the price of finding out, which is what it was always for.
The first mistake is renegotiating instead of deciding. A price reduction feels like a response to a finding and frequently is not: if the finding was that the asset is a different asset, a discount on the original asset does not address it, and the process continues with everyone's attention on a number.
The second is letting the walk-away note be rewritten by the person who is now tired. The note said materially concentrated; the week-fourteen version says materially concentrated by the standards of this industry. That amendment is not analysis, and it always arrives at exactly the moment it is needed.
The third is the audience. A great deal of what keeps a dead process alive is the cost of telling people it is dead - the partner who introduced it, the adviser who has worked on it, the seller who has been open with you. These are real relationships and real obligations, and none of them is a reason to commit capital.
The fourth is treating a walk-away as a failure of the process. A process that ends in a decision not to proceed has worked. It cost 8.0% of the contemplated amount and it prevented the other 92.0% from going somewhere it should not have gone, which is an excellent outcome and is almost never described as one.
The fifth is walking away without writing down why. The only thing that makes the next process cheaper is a record of what stopped this one and whether that judgement turned out to be right.
A price above which you do not proceed and two or three findings after which you do not proceed, in specific language. Date it, and keep the original text rather than a working copy. The value of the note is entirely that it was written by somebody who had spent nothing.
Would I enter this, on these terms, with these findings, if it arrived today and I had spent nothing? Say it to another person, because the version you say out loud is noticeably harder to fudge than the version you think.
60,000 spent against 750,000 at risk. Write both numbers on one line. The comparison is trivial once it is written and almost never happens, because the sunk cost operates as an atmosphere in the room rather than as a number on a page.
Tell the people who need telling, early, with the actual reason. The obligation is to be straightforward with them, and it is discharged by a conversation - not by committing capital. Conflating the two is how a social cost gets paid for with money.
At every checkpoint in a live process, and particularly the moment somebody proposes restructuring rather than repricing - that proposal is the reliable signal that the original thesis has stopped being true.
Walking away is not a virtue in itself, and a person who stops at the first serious finding will never own anything worth owning. Every real asset generates findings; the discipline is having written down in advance which ones end it, and then honouring the note in both directions - proceeding when the findings are inside your conditions is exactly as much a matter of discipline as stopping when they are not.
A sequence that repeats across private transactions: a finding emerges that the original thesis cannot absorb, and rather than ending the process the parties redesign the transaction around it - part of the price deferred, some of it contingent on the risk not materialising, a warranty attached. Everyone is acting in good faith and the new structure genuinely does reduce the exposure. What it does not do is restore the reason for proceeding, and because the discussion is now about mechanics, the question of whether this is an asset worth owning at all quietly stops being asked. The tell is that the thesis document is never revised, because revising it would require writing down what has replaced the original reason to invest.
When the response to a finding is a structure rather than a price, rewrite the thesis in one paragraph before agreeing to anything. If the new paragraph is not one you would have acted on in week one, the structure is keeping the process alive rather than the investment sound.
Create a free account to read the rest and apply it to your own investing.