Why Financial Resolutions Fail — and What to Build Instead

Around the second Friday of January, sometimes called “Quitter’s Day”, a large share of New Year’s resolutions quietly die. After nearly two decades in the financial industry, I can tell you this is not a personal failing. It is arithmetic.

Resolutions are, in a sense, the biggest thing we sell ourselves. And the people who manage serious money have quietly opted out of the entire ritual. Picture an investment committee overseeing tens of billions of euros, sitting down for its annual planning meeting. Nobody in that room says “this year, let’s save more.” Instead, they open a short written document, an Investment Policy Statement, that already says: if X happens, we do Y. No emotions. No resolutions. Just rules.

Outside those boardrooms, normal people resolve to save more and spend less. Guess which group is more successful.

Educational content only — not investment advice.

Willpower has a shelf life

The most-cited long-run research on this comes from Professor John Norcross at the University of Scranton, whose studies span from 1978 to 2020. His finding, repeated across the data: of the people who make a resolution, only about 46% are still succeeding six months later, and that is the group who kept going at all. Among comparable people who wanted the same change but set no resolution, the success rate is a single-digit percentage.

Notice what that means. Even a resolution is better than nothing, but by summer more than half of the committed have fallen off. A resolution is a wish about your future behaviour with no mechanism to enforce itself when motivation fades.

This is not weakness. Part of it is present bias, the well-documented way your brain values a reward now far more than a larger reward later. Ten euros today simply feels better than a hundred euros in a year. You cannot lecture that away. But you can engineer around it.

Your resolution has a technical name, and so does the thing that beats it

This is the part I wish someone had told me at twenty-five, because it is not motivational advice. It is a distinction from psychology that has been tested for three decades, and once you have it, the rest of this article stops being tips and starts being architecture.

Peter Gollwitzer and Paschal Sheeran call what you wrote on 1 January a goal intention. Their definition is dry and exact: goal intentions are “the instructions that people give themselves to perform particular behaviors or to achieve certain desired outcomes”, and they take the form “I intend to achieve X.”

Read that against your resolution. I intend to save more. I intend to stop panicking. It fits perfectly, which is the problem.

Because here is what the research found about goal intentions. Across a review of health-behaviour studies, people translated their good intentions into action only about 53% of the time. Not their vague hopes. Their genuine, sincere, measured intentions. A coin flip with slightly better manners.

Two kinds of intention

Why a resolution has no mechanism

Goal intention

"I intend to achieve X."

• · I will save more this year.
• · I will stop panic-selling.
• · I will invest regularly.

Names the destination. Says nothing about the moment of action.

Implementation intention

"If situation Y occurs, then I will do Z."

• · On the 1st, €400 moves to the broker automatically.
• · If a bucket drifts 5 points, I rebalance that week.
• · If the dividend is cut with no plan, I sell.

Names the cue and the response. The decision is already made.

Across reviews of health behaviour, people acted on their good intentions about 53% of the time. Not because the intentions were insincere. Because an intention is a wish about a future moment, and the moment arrives with its own weather.

The right-hand column is the entire rest of this series. A rebalancing threshold, a kill criterion, a standing transfer, a written exit plan: every one of them is an if-then plan wearing a financial hat, and every one of them was decided when nothing was happening.

The alternative has a name too. An implementation intention is, in Gollwitzer and Sheeran’s words, an “if-then plan that links situational cues (i.e., good opportunities to act, critical moments) with responses that are effective in attaining goals”. The form is fixed: “If situation Y is encountered, then I will initiate behaviour Z.”

That sounds like a small linguistic difference. It is not, and the reason it is not is mechanical rather than motivational.

A goal intention leaves the decision sitting in the future, to be made by a version of you who is tired, or frightened, or looking at a red screen. An implementation intention moves the decision to now, and hands the future version of you an instruction instead of a choice. Gollwitzer and Sheeran describe what happens then: the response specified in the if-then plan “becomes automated”, showing what they call immediacy, efficiency, and “redundancy of conscious intent.”

That last phrase is worth keeping. Redundancy of conscious intent. It means the plan works without you having to want it to, at the moment it fires. Which is exactly the moment your wanting is least reliable.

And notice what the institutional documents in this series actually are. An Investment Policy Statement is an if-then plan. A rebalancing threshold is an if-then plan. A kill criterion, a stop rule, a standing monthly transfer: all of them specify a cue and a response, both decided in advance, in calm weather. Nobody on an investment committee is more disciplined than you are. They have simply written down more of their decisions before the moment arrived.

The strongest evidence for this is about a savings plan, and it is brutal

If you want one study that settles the argument, it is not about resolutions at all.

In 2001, Brigitte Madrian and Dennis Shea published a paper in the Quarterly Journal of Economics about a large American company that changed one thing in its retirement plan. Before the change, employees had to actively sign up. After it, they were enrolled automatically unless they opted out. Nothing else moved. In their own words, “none of the economic features of the plan changed.”

Participation rose sharply. And more than that, the defaults stuck: a substantial share of employees hired under automatic enrolment simply kept the default contribution rate and the default fund, an outcome that almost nobody had chosen for themselves under the old system.

The paper’s title is the finding: The Power of Suggestion.

Think about what that means for every article you have ever read about financial discipline. The same people, with the same salaries, the same intelligence and the same intentions, saved dramatically differently depending on which way round the form was written. Not because anybody persuaded them. Because somebody changed the default.

You cannot restructure your employer’s pension paperwork this afternoon. You can change your own defaults, and that is the entire practical content of this article.

The money version of the same failure

There is a measurable version of willpower-failure in investing, and it is called the behaviour gap: the difference between what a fund returned and what the average euro invested in it actually earned.

It is real. Its size is genuinely disputed, and I would rather show you the argument than pick the scariest number in it.

How big is the behaviour gap?

Percentage points a year, given up to timing

DALBAR, for the year 2024 8.48

The number most often quoted. The same study put 2025 at 0.72.

DALBAR, for the year 2025 0.72

Same method, same publisher, one year later.

Morningstar, ten years to Dec 2024 1.2

Investors earned 7.0% where the funds returned 8.2%. Stable across periods.

The top two bars come from the same study, one year apart. A measurement that moves that far in twelve months is not measuring a stable human trait — it is measuring the market that happened.

And even the conservative figure is contested. In 2026 the Financial Analysts Journal published a paper by Fulkerson, Jordan, Riley and Yan whose title states its case plainly: bad timing does not cost investors 15% of their funds' returns. I have not been able to read the full paper, so I am not putting a rival number on this chart.

So in September 2026 I rebuilt the number myself, from 341,049 public SEC filings, with the data and code published alongside. Across 7,073 US funds the base-case gap for 2020 to 2025 is 0.36 points a year — and it moves between −0.06 and +0.62 depending on one share-class assumption that the famous figures never state. The full working paper is at /research/behaviour-gap.

What survives all of it is the direction in most measurements, and the size in none of them. Investors tend to do somewhat worse than the funds they own, by an amount that depends heavily on how you measure it. That is enough to act on, and the honest version is more useful than the frightening one.

The figure you will see quoted most often comes from DALBAR, and for 2024 it was 848 basis points. The same study, one year later, put it at 72. When a measurement moves by a factor of twelve in twelve months, it is telling you about the market that happened rather than about a durable human tendency. And it cannot be recomputed, because the method is not published. So I rebuilt it from public SEC filings: across 7,073 US funds the base-case gap for 2020 to 2025 is 0.36 points a year, and the range runs from −0.06 to +0.62 depending on one share-class assumption. Data and code are public — the paper is here.

Morningstar’s Mind the Gap is the more careful measurement: over the ten years to December 2024, the average dollar in US funds earned 7.0% a year while the funds themselves returned 8.2%, a gap of 1.2 percentage points that has stayed roughly the same across several ten-year windows. That is the number I use, and it is the conservative one.

Even it is now being argued with. In 2026 the Financial Analysts Journal published a paper by Fulkerson, Jordan, Riley and Yan making the case that the standard framing overstates the damage. I have not read the full paper and I am not going to pretend otherwise, so treat this as a live disagreement rather than a settled fact.

What none of them dispute is the direction. Investors do worse than the funds they own, and the cause is timing rather than fees or a bad market. A percentage point a year, given up to nothing but the moment. (The same measurement problem, and what it means for portfolio structure, comes up again in rules over access.)

Replace goals with systems

Here is what to build instead of resolutions. Read each one as an if-then plan, because that is what they are.

1. Automate the saving

Do not resolve to save. Make the money disappear before you see it. Set an automatic transfer on day one of the month, from your main account into your investments. The decision happens once, at a calm keyboard, not thirty times a month against temptation. This is Madrian and Shea’s finding applied to a household of one: change the default, not the person.

Pay your investment first. Treat it like rent, not like whatever’s left over.

(The full architecture — emergency fund, liquidity, growth — is the three-bucket system.)

2. Write one page

Not fifty pages. One. Your personal Investment Policy Statement says three things:

• Goal: e.g. a target amount by a target year.
• Risk: the loss you can genuinely sit through without selling, for many people something like a 30% drop.
• Rule: how the money is invested, and what would have to happen for that to change.

Then you write it once and don’t rewrite it every January. You revisit it only when something fundamental changes, such as a marriage, a child, an inheritance. While others rewrite their resolutions each year, you follow the same boring plan that actually works.

3. Stop waiting for the perfect moment

“I’ll start investing” fails because people wait for perfect conditions that never come. But the market has finished positive in roughly 73% of years since 1928, about three years in four. Waiting in cash “until it’s clearer” means fighting the base rate most of the time. (Why trying to pick the winners usually loses to just owning the market is here.)

4. Allocate forward, don’t track backward

Expense-tracking apps look in the rear-view mirror, and the money is already gone. Institutions use rolling forecasts: they assign money to purposes before the month starts. Rent bucket, food bucket, investing bucket, fun bucket. When the fun bucket is empty, fun’s over. No tracking, no guilt.

5. Then do nothing

This is the hardest and most valuable step. Once the plan is written and automated, leave it alone. No daily checking. No reacting to headlines. No optimising. Professionals even measure this, prizing a low activity ratio. The behaviour gap above exists precisely because people can’t stop tinkering. (When an exit genuinely is justified, and how to decide in advance, is here.)

The whole thing on a fridge door

Your resolutions will fail, and not because you are weak. They will fail because a goal intention is a sentence about the future, and a system is a decision already taken.

Amateurs have goals. Professionals have systems.

Take an hour today. Write one page: goal, risk, rule. Set up one automatic transfer. Turn off the notifications that tempt you to fiddle. Then act as though you’ve handed the keys to someone more patient than you, because in a sense you have.

Pick the one step from this list you can do today, and actually do it. One implemented action beats fifty watched videos — including mine.

Educational content only — not investment advice, and not a personal recommendation. Figures are historical and illustrative; past performance does not guarantee future results. Speak to a qualified, licensed professional before acting.