Suneet Singal on sunk costs

Suneet Singal's essay on sunk costs separates money, time, and effort already spent from the choice that is still open. The page sits beside the podcast The Opportunity Notebook with Suneet Singal and does not quote him. A sunk cost is a cost the present decision cannot recover. It can explain how you arrived here. It cannot, by itself, tell you whether to continue. Richard H. Thaler's work on mental accounting, described in the Nobel Prize materials for his 2017 prize in economic sciences, helps show why people sort losses into stories that feel unfinished. The unfinished feeling is real. The remaining decision still has to be judged by what happens from now, including the opportunity cost of staying. His published background in real estate and corporate finance is mentioned only as biography. This page gives no investment instruction. A labeled hypothetical uses a half-finished community project and invented figures.

A cost the decision cannot recover

A sunk cost is a cost you have already incurred and cannot get back, whatever you choose next. The money paid to a printer who does not give refunds, the weekend already spent, the nonrefundable deposit: these are facts about the past. They explain how the present situation was built. They do not change the comparison among the futures still available. The futures differ by what you will spend from now on and by what you will get from now on. Adding the sunk cost to every future, or to only one of them, does not help you see those differences. It is already the same number on every path, or it is a number that belongs to none of the paths because it is gone.

People resist this because it feels like the past expenditure is being insulted. It is not. The expenditure can have been reasonable when you made it, with the information you had then. The question on the table is different. Given that the expenditure is gone, which available future is better? Hammond, Keeney, and Raiffa name the sunk-cost trap in their Harvard Business Review article on hidden traps in decision making: the inclination to perpetuate the mistakes of the past. The trap is not caring about history. The trap is letting history cast a vote it cannot cast. A good past reason can sit beside a good present reason to stop. Both can be true.

Suneet Singal's published professional background is in real estate and corporate finance. Those fields are full of money already spent and choices still open. This page does not say how he treats any of them, and it does not tell a reader to buy, sell, hold, or abandon an investment. The tool is general. A household, a club, and a project can all ask which costs are already sunk and which costs are still choices. Write the sunk items in a column labeled past. Write the open items in a column labeled still ours to spend or to refuse. Only the second column belongs in the decision.

Mental accounts and an unfinished story

Richard H. Thaler received the 2017 prize in economic sciences for contributions to behavioural economics. The Nobel facts page summarizes the award. The popular background and the press release go further into one of his ideas, mental accounting. People simplify financial choices by creating separate accounts in their minds and judging a choice by its effect on a particular account rather than on the whole. A household might keep a holiday account and a bills account and refuse to let one repay the other. The Nobel materials note that this can help people protect a long-term plan, and that it can also lead them to borrow expensively while cash sits in another mental jar. Another part of the idea is the reference point. People judge outcomes relative to a reference, which can be the price they paid.

A sunk cost often lives inside an account that feels open until it shows a gain. We have spent $400 on the shelves, so we have to finish, or the $400 account shows a loss. The $400 is sunk if the supplier will not return it. Finishing spends more money and more weekends in order to make the mental account feel closed. Sometimes finishing is still the best future. Sometimes it is an expensive way to avoid writing loss on a jar that exists only in the story. Thaler's point, as the Nobel pages tell it, is that the jar is a psychological simplification, not a requirement of arithmetic. You can thank the jar for whatever discipline it gave you and still refuse to let it spend the next dollar.

Kahneman, Lovallo, and Sibony include a related check in their decision-quality article: are the people making the recommendation overly attached to past decisions? Attachment is the feeling the mental account produces. The check is to ask the question of a recommendation before you accept it. If the main reason to continue is that stopping would make the past look like a waste, you are inside the trap. Waste is a description of the past. The future is the weekends and the money you have not spent yet. Aim the decision at those.

Why the past outlay still feels like a reason

The feeling has several ordinary sources, and naming them makes them easier to set aside. One is consistency. Stopping looks like admitting that the earlier choice was wrong, and people prefer a story in which they were right all along. You can separate the stories. The earlier choice can have been right given what you knew, and the next choice can be to stop given what you know now. A second source is fairness to yourself. The weekend you gave up deserves a result. It deserves acknowledgment. It does not deserve a second weekend sacrificed so that the first one feels purchased. A third source is an audience. A board, a family, or a social feed heard you announce the project. Stopping in public is awkward. Awkwardness is a real social cost. Put it in the future column if you must, as the discomfort of an announcement, and keep the sunk money out of that column. You are allowed to count a real future embarrassment. You are not allowed to double-count a past invoice.

Hal R. Arkes and Catherine Blumer's paper The Psychology of Sunk Cost is a classic experimental account of the tendency to continue because of what has already been invested. This page cites it as that account. It does not invent a percentage from their studies, and it does not need one. The meeting-level observation is enough. People treat sunk costs as reasons. The remedy in the room is a rewritten comparison that includes only future costs and future effects. If a colleague says we have come too far to quit, ask them to translate the sentence into a future. Too far is not a place on next month's calendar. What will we spend next month, and what will we get, if we continue, and if we stop?

There is a kind version of the same question. People who poured themselves into a project are allowed to grieve a stop. Grief is not a math error. The error is making the next spend a monument to the grief. Give the grief an honest sentence in the notes, and then decide the future on the future's terms. Groups that mock the feeling will get the sunk cost back in disguise, because nobody will admit why they are voting to continue. Groups that honor the feeling and still separate the columns can stop without humiliating the people who did the early work.

The remaining decision

State the remaining decision in the form the first page of this set asks for. Who will commit, to what alternatives, by what act? The alternatives might be finish, stop and salvage, or pause for a month. The sunk cost is the same in all three, so it drops out of the comparison. What differs is the future spend, the future result, and the opportunity cost of each path. Opportunity cost is the next-best use of what you have not yet spent. If finishing takes two weekends that would otherwise revise the watering schedule, the watering schedule is part of the cost of finishing. It was never part of the sunk invoice. Henderson's definition, on the opportunity-cost page, is the tool. Use it on the resources that are still allocatable.

A pause is sometimes a way to avoid the feeling without making the decision. Pauses are legitimate when a dated fact would change the future comparison, such as a supplier's answer you do not yet have. A pause whose real purpose is to avoid saying stop is the sunk-cost trap on a calendar. Put a date on it. On that date, the same two columns reappear. If you find new reasons each month to extend the pause, write that pattern down. The pattern is information about the decider, not about the project.

Reversibility matters here. Money already spent is irreversible by definition if it cannot be recovered. The future path may still be reversible. You can finish one bay of shelves, look at whether anyone uses them, and only then spend the next bay. That trial does not resurrect the sunk money. It stops you from sinking more on the strength of a story. The reversible-decisions page is the companion. A sunk cost is a door that has already shut. It is not a reason to sprint through the next door because the first one hurts to look at.

A labeled hypothetical about shelves

This is a labeled hypothetical. The club, the prices, and the choices are invented. A community workshop has paid a cabinetmaker $400 to start a set of shelves. The figure is made up. The maker has done the cutting, does not refund the $400, and says finishing will cost another $250 and two further Saturdays of the club's labor. As of today, a member has found a used set of shelves that would serve the same wall, available for $90, a second invented figure, and installable in one Saturday. The remaining decision is whether to pay the extra $250 and two Saturdays, or to buy the used set and spend one Saturday, or to leave the wall empty and use the Saturdays on the watering schedule. The $400 is sunk in all three futures. It should appear in the notes as history and drop out of the comparison.

Compare the futures only. Finishing costs $250 plus two Saturdays and yields custom shelves. The used set costs $90 plus one Saturday and yields adequate shelves, in this story's own judgment. Empty wall costs no more cash and frees two Saturdays for the watering schedule, which the garden chair says is the work that prevented last summer's losses. The sunk-cost sentence we already paid $400, so we must finish does not distinguish these futures. It merely protects a mental account in which $400 must be followed by a finished custom wall or else count as waste. The account is understandable. It is not one of the futures.

The hypothetical refuses a winner on purpose, because different clubs value custom work differently. What it does not allow is a winner chosen by the $400. If the club values custom shelves enough to spend $250 and two Saturdays instead of $90 and one, it should say that in the future tense. If it would not buy the custom job today as a fresh choice, the $400 is not a reason to buy it as a continuation. Nothing in the story is an investment tip, a woodworking quote, or an event from Suneet Singal's life. The numbers exist so the sunk column and the future column cannot be confused.

Learning is not the same as clinging

Stopping is not the only way to respect a sunk cost. Sometimes the remaining future is still the best future, and the right sentence is we continue, and the past spend is not why. A half-built shelf that will be finished for $250 and will do a job the used set cannot do is a fine thing to finish. The test is whether you would accept the future spend if a stranger had paid the $400. If yes, continue. If you would only continue because the $400 was yours, the ownership of the past is doing the work, and Thaler's reference point is in the chair. Ask the stranger version out loud. It is a simple way to move the reference point from the price you paid to the choice you still have.

Learning from the past is a third column, separate from both sunk cost and future spend. We learned that this maker requires full payment before a refund window. We will write the next agreement differently. That lesson is valuable and free of the trap, because it changes a future agreement rather than justifying the current continuation. Do not confuse the lesson with a reason to throw more money at the same agreement so that the lesson feels earned. The lesson is already earned. Additional spending needs its own future benefits.

There is also a base-rate version of the lesson, taken up on another page. If projects of this kind often stall at the halfway mark, the stall is not a personal failure that more money must redeem. It is a feature of the class. Knowing the class does not tell you the fate of this shelf. It stops you from treating this shelf as a unique moral drama in which quitting is a special disgrace. Most dramas of that kind are the sunk-cost feeling wearing a costume. Take the costume off, keep the lesson, and decide the remainder.

A sentence that keeps the past in its place

Use one sentence when the conversation slides. The money and the time already spent are sunk. Here is the comparison of what we would still have to give, and what we would still get. Put the sentence at the top of the memo. If a speaker returns to the sunk amount, point to the sentence and ask for a future difference. Do this without contempt. The speaker is usually protecting work they were proud of. The sentence protects them from spending more pride after it. Kahneman, Lovallo, and Sibony suggest that reviewers look for attachment to past decisions. Be that reviewer even if the team is you, by writing the sentence the night before and reading it when the attachment is loud.

Keep sunk costs out of expected-value sheets. A sheet that adds the $400 to the cost of finishing, and not to the cost of switching, has cooked the ranking. If you insist on showing the $400, add it to every option, including stopping, and watch it cancel out of the comparison. Canceling out is the arithmetic form of the lesson. If it does not cancel, you have built the sheet to scold one of the options for the past. Rebuild the sheet. The expected-value page is about future outcomes and their weights. The past is not an outcome you can still weight.

Then choose, and say the choice in the future tense. We will finish because the next $250 is worth the custom fit. Or we will switch because one Saturday and $90 meet the need. Or we will stop and give the Saturdays to the garden. Any of those can be right. We will finish because we already spent $400 cannot. The past can be honored in a different sentence, spoken once, and left in the archive. The decision gets the future. That split is the entire content of a sane attitude toward sunk costs, and it is the split this page asks you to write down before the next dollar or the next Saturday leaves your hands.

What this entry does not claim

Suneet Singal’s published professional background is in real estate and corporate finance. Company filings linked from this site record a chief executive officer departure from FC Global Realty Incorporated effective 2 January 2018 and an acting chief executive officer appointment at First Capital Investment Corporation effective 12 September 2017. The Suneet Singal Scholarship, Grant and Giving websites describe him as the founder of those programs. This entry explains a general subject. It is not a private track record, a current job title, an investment result, or a quote.

A sunk cost is history. The live alternative is on opportunity cost, the door you have not yet shut is on reversible and irreversible decisions, and the scoreboard you may be protecting is on whose definition of success you are using.