Here’s a question I keep coming back to. If you can buy your way out of almost every problem a marriage is supposed to have, why do the marriages that bought their way out still fall apart?

I’m thinking about a very specific kind of marriage, because it’s the one I actually know: two people in a city like Bangalore, both working, a household income that clears a crore, a cook, a maid who comes twice a day, a driver, a nanny once there are kids. On paper, look at what that setup deletes. Money stress? Cushioned. Who does the dishes? Nobody in this house does the dishes. That entire Western marriage-advice industry about splitting the chores fairly is, for us, mostly a solved problem. We paid someone.

And yet.

So I went looking for what the research actually says, and the first thing that falls over is money.

You’d assume that above some income the money fights just stop. They don’t. When you control for what a couple actually earns and owes, the amount of money barely predicts anything. It’s the fights about money that predict divorce, not the money.¹ Sit with how strange that is. A couple pulling a crore can have the exact same corrosive argument as a couple pulling five lakh, because the argument was never really about affordability.

Then what is it about? When researchers coded what couples actually say when they fight about money, the top themes weren’t “we can’t afford this.” They were things like you’re being irresponsible, this isn’t fair, I contribute more than you do.² Money is just the surface. Underneath it is a question about respect, and about whether the two of you are a team or two people quietly keeping score.

That keeping-score thing turns out to be close to the whole game, and there’s a genuinely wild piece of evidence for it. A few years ago a group of researchers pooled 43 separate long-term studies, something like eleven thousand couples, and pointed machine learning at the question of what actually predicts whether a relationship is any good. The winner wasn’t personality. It wasn’t how “compatible” two people looked on paper. The strongest predictors were all relationship-specific feelings: do I feel appreciated, do I feel my partner is committed, do I feel responded to. How your partner makes you feel beat who your partner is, and it wasn’t close.³ Your partner’s own traits added almost nothing once you knew how the other person felt.

Which brings me to the thing I actually wanted to write about. The maid.

Here’s the trap for people like us. We look at the chore research, notice we’ve outsourced the chores, and conclude we’ve bought our way out of the fairness problem. We haven’t. We’ve just moved it somewhere we can’t see.

Because the thing that stuck to one person was never the scrubbing. It’s the remembering. Somebody still has to notice the maid didn’t show, reorder what ran out, know the kid’s vaccination is due, decide what’s for dinner before anyone’s hungry, hold the running list of everything the household needs in their head. Sociologists have a name for this now: cognitive labor. And study after study finds it lands disproportionately on the woman, and that it’s uniquely draining precisely because it’s invisible.⁴ You can’t watch someone remember.

And here’s the finding that should end the “but we have help” defense. Somebody actually checked whether money buys you out of this. It doesn’t. Higher-earning mothers do less of the physical work, sure. But they carry the same mental load as everyone else. The planning and the worrying don’t transfer to paid help, because they require knowing your own family, and that knowledge isn’t for sale.⁵

Even hiring more help doesn’t cleanly fix it. Bring in full-time or live-in help and a chunk of the time you save gets eaten by a new job: managing the help. Recruiting, training, scheduling, correcting, smoothing things over. The labor doesn’t vanish, it shape-shifts, from doing into supervising, and the supervising stays stuck to the same person.⁶

So the fight in an affluent home was never going to be about the dishes. It was always going to be about whether the invisible work gets seen. Whether the person carrying the running list in their head feels like the other one even knows it’s there.

There’s a flip side to being seen, and it’s the thing that actually kills marriages. Gottman’s people like to say contempt is the single biggest predictor of divorce: the eye-roll, the little sneer, the sense that your partner has started to look down on you. To be honest, the famous prediction numbers around this are shakier than they get quoted as. Someone went and showed that the “we can predict divorce with ninety-odd percent accuracy” models were basically fit after the outcomes were already known, which is a much weaker claim than it sounds.⁷ But strip the inflated stat away and the direction survives. Contempt is the exact opposite of feeling seen. It’s one person broadcasting I don’t respect your inner world.⁸ That’s the poison.

Now the part where I have to be careful, because almost everything above was measured on Western couples, and I’d be lying if I said it maps cleanly onto us. Most of this research comes out of the single weirdest slice of humanity to generalize from: Western, educated, rich, and studied to death precisely because it’s easy to study.⁹ So what actually survives the trip to a marriage in Bangalore?

Two honest things. First, the one piece of this that has been checked on Indian couples holds up: feeling understood and validated really does separate the happy urban Indian marriages from the unhappy ones,¹⁰ and feeling your partner is responsive predicts satisfaction here just like it does anywhere.¹¹ The core survives translation.

But the form it takes might not. There’s a study where the standard Western advice, respond to your partner’s good news with big visible enthusiasm, actually backfired for South Asians. The understated response, the quiet acknowledgement, landed better.¹² Which tracks, doesn’t it? Empathy here has always run more through what you do than what you announce. Showing up. Handling the thing before you’re asked. The love language is closer to seva than to a Hallmark card.

And the honest bottom of it: nobody has really studied couples like us. There’s one decent survey of dual-career couples in Bangalore, and even that one isn’t about the two-crore, help-employing, both-in-tech household.¹³ I’m stitching together Western data, a bit of pan-India data, and my own inference, and I want to be upfront that the stitching is mine. Even our beloved “Indians just don’t divorce” stability is mostly an illusion of constraint rather than a report of happiness. The divorce rate is among the lowest on earth, but separations quietly outnumber divorces several times over, and a low exit rate mostly measures how hard it is to leave.¹⁴

So where does that leave a couple who did everything right on paper?

Here’s my read, for whatever it’s worth. Money is a fantastic solvent. It dissolves the scarcity fights, it dissolves the chore fights, it buys back time. And that is exactly the trap, because it dissolves all the legible problems and leaves you standing in front of the one that has no price. Whether the person across the table feels seen. Whether the invisible work registers. Whether, when they tell you something good, you make them feel it landed.

You can’t outsource that to a maid, or expense it, or solve it with a bigger house. It’s the oldest and cheapest thing in the world, and somehow it’s the only part left that’s genuinely hard. Maybe that’s the point. Once you’ve bought your way out of everything money can fix, what’s left standing is the part that was ever really the marriage.

I’m still figuring out how to get better at it. But at least now I know that’s the thing to get better at.

References

Footnotes

1. Dew, Britt & Huston (2012). Examining the Relationship Between Financial Issues and Divorce, Family Relations 61(5), 615-628. Financial disagreements were the strongest disagreement-type predictor of divorce, and once they were in the model, financial well-being (assets, debt, income) no longer predicted divorce. ↩

2. Peetz, Meloff & Royle (2023). When couples fight about money, what do they fight about?, PLOS ONE. Dominant conflict themes were perceived irresponsibility, unfair relative contributions, and fairness, not scarcity; framing a disagreement as unfairness predicted worse outcomes. ↩

3. Joel, Eastwick, et al. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies, PNAS 117(32), 19061-19071. Across ~11,000 couples, relationship-specific perceptions (felt appreciation, commitment, responsiveness) predicted relationship quality far better than individual traits, and a partner’s own traits added little beyond the focal person’s felt experience. ↩

4. Daminger (2019). The Cognitive Dimension of Household Labor, American Sociological Review 84(4), 609-633. Anticipating and monitoring household needs is a distinct, taxing, largely invisible form of labor that skews female. (US, college-educated sample.) ↩

5. Weeks, Kowalewska & Ruppanner (2025). Take a Load Off? Not for Mothers: Gender, Cognitive Labor, and the Limits of Time and Money, Socius. Higher-earning mothers did less physical labor but reported the same mental-load tasks: money did not buy out the cognitive load. (US sample; the direction is the point, the exact magnitudes in the paper are still worth checking.) ↩

6. Cheung & Lui (2024). Outsourcing domestic work: A double-edged sword for marital relations among dual-earner couples, Family Relations. Outsourcing eases role strain but also creates new conflict over managing the help and evaluating each partner’s contribution. (Hong Kong dual-earner sample, the closest large study in a familistic Asian setting, not India.) ↩

7. Heyman & Smith Slep (2001). The Hazards of Predicting Divorce Without Crossvalidation, Journal of Marriage and Family 63(2), 473-479. The famous 90%+ divorce-prediction accuracies were fit to already-known outcomes and never cross-validated; on fresh data the predictive value collapses. ↩

8. Lisitsa (2013). The Four Horsemen: Contempt, The Gottman Institute. The Gottman Institute’s own framing of contempt as the strongest divorce predictor; a proponent source, so treat the ranking as their claim, but the descriptive pattern (contempt as corrosive disrespect) is well attested. ↩

9. Henrich, Heine & Norenzayan (2010). The weirdest people in the world?, Behavioral and Brain Sciences 33(2-3), 61-83. Western, Educated, Industrialized, Rich, Democratic samples are frequent global outliers, so psychology findings generalize less than their phrasing implies. ↩

10. Sandhya (2009). The Social Context of Marital Happiness in Urban Indian Couples: Interplay of Intimacy and Conflict, Journal of Marital and Family Therapy 35(1). In 91 urban Hindu couples spanning arranged and love marriages, happy couples reported more empathy, validation and support. ↩

11. Thargay, Giri & Behera (2026). Feeling good to feeling satisfied: The mediating role of perceived partner responsiveness, Journal of Social and Personal Relationships 43(8). In 203 married couples in Sikkim, one’s own positive affect predicted marital satisfaction partly through perceiving the partner as responsive. ↩

12. Cross-cultural differences in supportive responses to positive event disclosure (2024), The Journal of Positive Psychology 19(2), 277-290. Across a combined sample of European Americans, East Asians and South Asians, South Asians who received passive-constructive (understated) responses to good news reported higher perceived responsiveness and satisfaction, the opposite of European Americans. ↩

13. Joseph & Inbanathan (2016). Marital Disharmony among Working Couples in Urban India, ISEC Working Paper 373, Bangalore. Among 238 dual-career Bangalore couples, “lack of time for each other” was a major conflict trigger reported far more by women than men. (Single-respondent, broad-professional sample, not the top-1% help-employing household.) ↩

14. Jacob & Chattopadhyay (2016). Marriage Dissolution in India: Evidence from Census 2011, Economic & Political Weekly 51(33). India’s divorce rate is among the world’s lowest (~0.24% of the ever-married), and the number of separated people is roughly three times the number divorced. ↩