Part V · Chapter 10 · Behaviour · lesson 1
Why the future self keeps losing
11 min
Builds on Interest, from both sides of the table
Part V · Chapter 10 · Behaviour · lesson 1
11 min
Builds on Interest, from both sides of the table
Two questions, before any explanation. Answer them the way you would actually answer them, not the way a sensible person would.
What would you do
Second question. Same person, same money, same seven days.
What would you do
A steady impatience would apply the same discount to every week — the week starting today, the week starting a year from today, all of them. That is not what people do. The discount is enormous for the week that starts now and shrinks fast for weeks further out.
The consequence is that preferences flip as time passes. In January you genuinely prefer the larger payment in December. When December arrives, the same person genuinely prefers the smaller payment today. Neither self is lying. They are the same discount curve, read from two positions on it.
This is worth being precise about, because the everyday word for it — willpower — points at the wrong thing. The problem is not that the future self is weak. The problem is that the future self is not in the room.
Every plan you make for next month is made by someone with no cravings, no bad day at work, no rain, and no fatigue. That person is a genuinely different decision-maker from the one who has to carry it out, and the planner keeps losing because the planner is never present at the moment of the decision.
It is tempting to file this as a defect. It is not one, or at least it did not start as one.
A promise about next year is a promise made by a world that might not deliver. For most of the time anything resembling a human has existed, food spoiled, seasons failed, and the person who owed you something moved away or died. Discounting the distant future steeply was not a bias in that environment — it was an accurate read of how likely the distant future was to pay out.
The heuristic did not stop working because it became stupid. It stopped working because the environment changed underneath it.
Reliable institutions, enforceable contracts and long lifespans made the distant future far more likely to actually arrive. The discount curve did not get the memo. It is still pricing delivery risk that a bank account has already removed.
Check
Not the same thing
Preferring sooner to later, at a consistent rate, however far away the choice is.
Applying a much steeper discount to the immediate period than to any later one.
Which is which? Put each one on a side.
Signing up for a course starting in three months and not attending
Refusing a savings product at any rate, in any year, on principle
Deciding on Sunday to cook all week, and ordering food on Tuesday
Two things, and neither of them is "try harder".
The first is that a decision made in advance and a decision made in the moment are different decisions, taken by different versions of you, and the one made in advance is usually the one you would endorse afterwards. That is an argument for moving decisions earlier, not for exerting more effort later.
The second is that anything with the shape benefit now, cost later is riding the curve, and anything with the shape cost now, benefit later is fighting it. Almost every accumulation of wealth has the second shape. So does exercise, so does study, and so does every unglamorous maintenance task in a life. It is not a coincidence that those are the things people find hard.
People who know their own curve sometimes buy constraints deliberately: paying in advance, telling someone else the plan, putting the thing on a schedule that is awkward to change. Each of these takes a future decision away from the future self, which is the point.
Constraints that carry a real cost for breaking them tend to hold better than constraints that do not, which is a very slightly depressing thing to know about oneself. It is also the design principle behind most of the systems in the last lesson of this track.