Years From Now, You’ll Wish You Had Given It Ten Years

One evening, ten years from now, you open an old file. Perhaps it contains unfinished research notes, an investment account you emptied midway through a plan, a skill you stopped learning, or a project you put aside after a year.

What will you think of the person you were then?

I suspect it will rarely be, “Thank goodness I gave that up.” More likely: “What might this have become if I had stayed with it for a few more years?”

That possibility is the starting point of this essay, because it suggests something an economist cannot easily ignore: some future regret can be anticipated.

In finance, a foreseeable future cost belongs in today’s decision. Yet most people price the difficulty they face now without pricing the regret they may feel later. That is a systematic error in valuation, and it may take years to become visible. By then, the original decision can no longer be revised.

The question I want to explore is simple: when you look back, you may wish you had chosen the long-term version of what mattered.

Why does that happen? Which pursuits deserve that commitment, and which do not? Let us work through it.

1. Regret changes with time

Start with a widely cited study. In their 1995 review in Psychological Review, psychologists Thomas Gilovich and Victoria Medvec described a temporal pattern in regret.

In the short term, people tend to regret actions that went wrong. Over longer periods, more of their regret concerns things they never did.

Later work offers related evidence. In their 2005 study in Personality and Social Psychology Bulletin, Neal Roese and Amy Summerville found that prominent life regrets cluster in areas such as education, careers, and relationships, often around missed opportunities.

The limitations matter. First, much of this research relies on self-reports and recollection, which can idealize and reconstruct the past. Second, perceived opportunity plays a role: Roese and Summerville found greater reported regret in domains where people still saw opportunities for improvement. Their result should not be reduced to the claim that regret is always strongest once an opportunity has closed. So I am not telling you that taking action guarantees freedom from regret. That would be a motivational slogan.

What I want you to consider is the structure beneath the idea.

The cost of an action can be bounded. If it goes wrong, you can see what happened, and the pain may dull over time.

The imagined cost of inaction is harder to bound. You never discover where the unchosen path would have led, and your imagination can keep raising its value.

In financial terms, short-term regret resembles a loss with a ceiling. Long-term regret can resemble regret over a forgone option: a call you might have held, but for which you chose not to pay the premium.

The long-term version matters because it usually requires extending the holding period, giving that possibility enough time to develop.

2. The short-term and long-term versions are different pursuits

This is the central point of the essay, and the one most easily mistaken for motivational advice. So let me be precise.

Most people understand a long-term approach as doing the same thing for a little longer. That understanding is incomplete.

The same activity can become a different pursuit at a different time horizon, because the objective you are optimizing changes.

Consider a few examples.

Investing. The short-term version optimizes how an account’s return chart looks this quarter. The long-term version asks whether your assets will remain productive a decade from now, with a base on which returns can compound. The former encourages chasing themes, frequent rebalancing, and avoiding any period that makes the account look bad. The latter directs attention to whether the businesses you own can keep creating value, and whether you can endure interim drawdowns without being forced out. The information, actions, mindset, and even stock-selection criteria differ.

Writing and publishing. The short-term version optimizes the views on a single piece. The long-term version asks whether readers will still recognize and trust you three years from now. The first rewards headlines and emotion; the second rewards an accumulation of judgment and credibility.

Learning. The short-term version asks how many tools you mastered this week. The long-term version asks whether, in ten years, you will have a framework that transfers across problems.

These are differences in the rules of the game, rather than in intensity alone. The short game rewards display, speed, and immediate feedback. The long game rewards options, trust, and a base for compounding.

That is why the usual advice to be patient and keep going misses the deeper question. Patience is a means. The real issue is:

Which game are you playing?

Many people believe they are playing a long game while scoring themselves by short-term rules: checking portfolio values daily, checking metrics weekly, changing direction monthly. They lose to the scoreboard they have chosen, rather than to time itself.

Some forms of value exist only over a long horizon. Reputation, expertise that is difficult to copy, a foundation for compounding, and deep trust between people are not merely small units of short-term value added together. They emerge after a threshold of time has been crossed. Compress them into a short-term objective, and you may end up with something different altogether.

3. The mind favors now; your future self gets the bill

If the long-term version is so valuable, why is it difficult to choose?

Behavioral economics offers one explanation: hyperbolic discounting. In the textbook model of exponential discounting, each additional period of waiting receives the same proportional discount. Human preferences often behave differently. We are unusually sensitive to the difference between now and soon, while barely noticing the difference between the distant future and a slightly more distant future. George Ainslie and others described this pattern, and David Laibson incorporated it into an economic model in his 1997 paper in the Quarterly Journal of Economics. The consequence is preference reversal: today we decide to begin tomorrow; tomorrow we decide to begin the day after.

The environment reinforces it. Modern society is a machine for immediate feedback: likes, performance targets, trending topics, daily portfolio values, and live returns. Each trains us to give greater weight to now than to later.

Where do the costs appear? Two bodies of evidence are worth considering.

First, Brad Barber and Terrance Odean’s Trading Is Hazardous to Your Wealth, published in the Journal of Finance in 2000, analyzed household accounts at a large US brokerage from 1991 to 1996. The most active traders earned a net annualized return of about 11.4%, compared with about 17.9% for the market. Trading costs associated with high turnover were an important source of the gap.

Second, Morningstar’s recurring Mind the Gap research finds that investors’ dollar-weighted returns have often fallen short of the returns of the funds they owned, reflecting the timing of purchases and sales. The precise gap varies by year and methodology. I recommend reading the latest report directly rather than treating my summary as a substitute.

Both point toward a similar lesson: how an investment is held can change the return its owner receives. Part of the difference comes from our behavior, rather than from the market alone.

My conclusion is that the short-term version is rarely cheap. It sends the bill to your future self.

The delay is what makes that bill deceptive. The patience you avoid paying for today may return in five or ten years, with interest, as the thought: “If only I had…”

4. Compounding needs something to keep compounding

Compounding is perhaps the most famous and most misused idea in this discussion. People call it the eighth wonder of the world, yet often overlook its mathematical shape:

Much of the gain arrives toward the later part of the curve.

Here is an illustrative calculation. These figures describe a hypothetical mathematical relationship; they are not a forecast or a promised return for any asset.

Suppose capital compounds at 8% a year:

• After ten years, it is worth about 2.16 times the initial principal.
• After twenty years, it is worth about 4.66 times the initial principal.
• The total gain is about 3.66 times the original amount. Roughly 68% of that gain comes in the second decade, and 32% in the first.

On this curve, the first half of the period produces the smaller share of the result; the second half produces most of it.

If you stop the growth after year ten and allow neither the principal nor its accumulated return to earn anything further, the value stays near 2.16 times the original amount. Continue under the same assumptions, and it reaches roughly 4.66 times by year twenty. Starting over here means abandoning what has accumulated. Switching accounts or investments does not, by itself, reset compounding: if the full proceeds continue earning the same return, the mathematics is uninterrupted.

That is what I mean when I say compounding punishes interruption more than a lack of intensity. The costly mistake is discarding your accumulated base during a period when its value is still easy to overlook.

Evidence from markets illustrates the concentration of returns. In his 2018 study in the Journal of Financial Economics, Hendrik Bessembinder found that fewer than 5% of US stocks accounted for the market’s entire net wealth creation from 1926 to 2016; the remainder collectively matched Treasury bills. Returns were concentrated in relatively few companies. A long holding period can help gains accumulate, but holding any particular stock for long enough does not guarantee a profit.

Warren Buffett is another familiar illustration: most of his wealth accumulated after he turned 50, a pattern discussed by several writers who have examined his wealth over time. The exact proportion changes with share prices.

But the story needs a warning about survivorship bias. We see the people who stayed in the game and chose well. Many others persisted just as long in the wrong assets or directions. They seldom feature in the story.

Time is an amplifier. It can magnify a sound direction, and it can magnify a bad one.

That changes how we should think about opportunity cost. Most people compare doing one thing now with doing something else now. That is a static comparison. A more useful one compares the eventual value of two paths: a pursuit continued for ten years, and one repeatedly abandoned and restarted every two. Starting again can be expensive when it means discarding the trust, skill, or relationships you have already built.

5. When AI makes speed cheap, depth matters more

Investing, writing, and personal development share this logic. AI can sharpen it, rather than make it less relevant.

The intuitive response is often: “If AI is this fast, does long-term accumulation still matter?” My judgment is the opposite.

Consider what AI changes. It drives down the cost of execution. Drafts, spreadsheets, information searches, and code that once required considerable time can now often be produced in minutes. Speed becomes cheap.

What becomes more valuable as speed grows cheaper? Judgment, taste, domain depth, and credibility. None arrives through a single prompt. They develop over years.

I find multiplication a useful way to think about this. AI is the multiplier; what you have built is the base being multiplied. Give the same tool to someone with a decade of accumulated expertise and to a beginner, and the results may differ dramatically. A stronger base gives the multiplier more to amplify. If the base is zero, multiplying it changes nothing.

Another way to frame it is that short-lived capabilities are depreciating while enduring ones become more valuable. You may spend three months learning a tool, only to find a new version has replaced it a year later. Such tool-specific skills have an increasingly short shelf life. Judging whether a business model can make money, whether a source is credible, or whether a question is worth asking can remain useful for decades.

Nassim Nicholas Taleb has popularized the Lindy effect: for certain nonperishable things, such as ideas, books, and technical standards, a longer history of survival can suggest a longer remaining life. It is a heuristic, not an iron law, and it does not apply to perishable things. As a filter, though, it can be useful: give some priority to what time has already tested.

My advice for the AI era is therefore to let AI handle work that depends on speed, while you do the work that depends on years. Invest the time it saves in assets whose value only time can realize.

6. Few things deserve ten years: four questions and a boundary

You might now ask whether everything deserves a ten-year commitment.

Quite the reverse. Few things do.

The cost of the long-term version is substantial: it occupies your scarcest resource, time. Committing years to one pursuit excludes many alternatives. Before deciding to make that commitment, I use four questions.

First: does it compound? Will today’s effort make tomorrow’s output easier? The trust accumulated through writing can do that; a one-off job may not.

Second: does it create a barrier others will struggle to cross? After ten years, would it be difficult for someone else to catch up in one? If anyone can replicate the result in a year, it may not deserve ten years of your life.

Third: does it expand your options? A good long-term investment gives you more paths to choose from, rather than gradually narrowing them.

Fourth: would I still do it without an external reward? This is the most honest question. A long commitment inevitably includes periods with little visible payoff. If the reward is your only reason for continuing, you may not last long enough to receive it.

I look for a yes to at least three of these questions before committing years.

Then comes the essential boundary: a long-term approach is different from the sunk-cost fallacy.

The sunk-cost effect, classically examined in Arkes and Blumer’s 1985 study, describes the tendency to keep pursuing something because much has already been invested, even when its prospects have deteriorated. On the surface, it resembles persistence: both tell us not to give up easily.

The difference lies in the reason for the decision:

• Sunk-cost reasoning looks backward: I have spent too much to let it go to waste.
• A long-term approach looks forward: the prospects still justify another commitment.

One question separates them: “If I were starting from zero today, would I choose this again?” If yes, the amount already spent is beside the point; you continue because the pursuit remains worthwhile. If no, even five years of prior effort should not oblige you to spend more of your future defending a past that no longer holds up.

Someone who takes the long term seriously can therefore be unusually decisive about stopping. They choose carefully before committing, and give a sound choice enough time to work.

Be selective before you begin. Be patient once the choice continues to earn that patience.

7. Return to your future self

Return to the evening at the beginning.

Ten years from now, you open that old file. You are unlikely to dwell on one particular portfolio drawdown, or remember how many people read a single essay. The question that remains is simpler:

Did you give the worthwhile pursuit its long-term version?

Let me put it plainly. A long-term approach is more than a temperament, an exercise in willpower, or a comforting story. It is an economic judgment:

A repricing of time, compounding, and opportunity cost. What needs to change is the scale you use to assess your actions, rather than the strength of your self-discipline alone.

I will leave you with three questions. You need not answer immediately, but take them seriously:

1. Which pursuit, if continued for ten years, would make your future self grateful? Are you currently approaching it as a short-term project?
2. What do you keep restarting? Each time you begin again, which part of your accumulated base do you discard?
3. If you were starting from zero today, would you choose your current path? If you would, perhaps what it still needs is time.

The short-term version is rarely cheap. It sends the bill to your future self. Every decision you make today either pays part of that bill in advance or leaves more of it outstanding.