The trait everybody gets praised for is the one the earnings data treats worst.
Agreeableness is the trait most likely to turn up in your performance review as a compliment. It is also the one the earnings data treats worst.
Two things get confused here immediately, so let us separate them before anything else. Agreeableness, as this test measures it, is a disposition: how readily you accommodate other people, assume good intent, avoid friction, go along with things. Kindness is a choice you make. They overlap, and they are not the same. Everything below is about the first one, and nothing in this literature says that being decent to people costs you money.
There are two findings here with two different outcome variables, and they get mashed together constantly. On income: Judge, Livingston and Hurst (2012), in the Journal of Personality and Social Psychology, found agreeableness negatively related to earnings. Their own summary is blunter than anything we would write — “‘Niceness’ — in the form of agreeableness — does not appear to pay.”
On financial hardship: Matz and Gladstone (2018), same journal, found highly agreeable people more likely to hold lower credit ratings, more debt and less savings, and to be more prone to bankruptcy. Not lower income — worse financial outcomes at whatever income they had. Both findings replicate, and both are small in the way nearly everything in personality research is small. We will put the actual number on that at the end, because it is the most important thing on this page.
The obvious story is that agreeable people are bad negotiators: they do not push, they take the first offer, they will not ask. It is a satisfying explanation, and Matz and Gladstone tested it directly.
It was not the answer. What predicted the financial difficulty was something quieter: agreeable people place a lower value on money. Not that they cannot negotiate — that the whole subject matters less to them. And a subject that matters less gets less attention, less tracking, and fewer of the small dull decisions that keep a financial life in order.
That is a considerably kinder finding than the folk version, and a more useful one. It does not say you are weak in a room. It says money is not the thing you are pointed at — which for plenty of people is a value rather than a defect, right up until the consequences arrive in the form of a credit score.
One more part of that study matters more than its headline. The effect was stronger, and sometimes only present at all, among people on lower incomes. With plenty of slack, inattention costs you optimisation. Without it, the same inattention costs you the rent. The trait bites hardest exactly where it can least be afforded, which is the opposite of how this finding usually gets reported.
The income penalty in the 2012 study was larger for men, and largest for older men. The authors read it as agreeableness sitting worse against what is expected of men at work, so men pay more for having it.
The obvious conclusion — that women have less to gain from becoming less agreeable — is half right, and the other half matters. A separate, well-replicated literature on backlash finds that women who assertively advocate for themselves are evaluated more harshly than men doing exactly the same thing: rated more competent, less likeable, and penalised materially for it. So “be less accommodating” is advice with a different price depending on who takes it.
There is one documented way around that, and it is oddly specific. The backlash largely disappears when women negotiate on behalf of someone else rather than themselves. Framing a request as advocacy — for the role, the team, the standard, the budget — rather than as a personal ask is not a trick. It is the version of the same conversation that does not trigger the penalty.
Not “be less nice”. That is bad advice and it is not what the research says. The useful move is to separate the part of agreeableness you want to keep from the part quietly costing you.
Everything on that list leaves you exactly as warm as you were. That is the point. The trait is not the problem; the unattended admin is.
The size of it. Across a meta-analysis of 63 studies and 896 effect sizes, the strongest of the five traits accounts for roughly 1.6% of the variation in earnings — and agreeableness is not the strongest. Education, occupation, hours, cognitive ability, whether you ever ask, geography and who your parents were do almost all of the work. Anyone converting an agreeableness score into a salary forecast is inventing it.
This test also cannot see the thing that would matter most: whether your accommodation is chosen or compulsive. Someone who gives way because they genuinely do not mind and someone who gives way because the alternative feels unbearable answer these ten statements identically, and only one of them has a problem. The score describes the behaviour, not your relationship with it.
It is not a hiring instrument either. Self-report measures are trivially gameable by anyone who wants the job, which makes them weak at selection — and legally riskier than employers tend to assume: a US appeals court has held that a clinical personality inventory counted as a “medical examination” under the Americans with Disabilities Act (Karraker v. Rent-A-Center, 7th Circuit, 2005). That case concerned a clinical test rather than this one, and we are not lawyers — but it is a reason to take advice rather than our word for it.
And if any of this landed uncomfortably, the useful question is not whether the research is right. It is which of the ten agreeableness statements you hesitated over.