Doing money beats knowing money
I write about money on the internet, which means my whole job rests on an assumption: that knowing more leads to doing better. A 2021 study of lower-income households in Malaysia tested that assumption directly, and I have some uncomfortable news about my industry. Knowledge came third.
(Rahman, Isa, Masud, Sarker and Chowdhury, Future Business Journal 7:52, 2021)
TL;DR: Among 412 bottom-40%-income households, financial behaviour was the strongest predictor of financial well-being (β = 0.333), financial stress the strongest negative (β = −0.256), and financial literacy the weakest of the three (β = 0.165). Literacy had no significant effect on behaviour at all. Behave, then de-stress, then learn, in that order.
The study
The researchers surveyed 412 households in Malaysia's B40 group, the bottom 40% by income, in the Kuala Lumpur metro area, and built a structural path model connecting three inputs (literacy, behaviour, stress) to financial well-being. Survey-based and cross-sectional, so hold that thought for the caveats.
The scoreboard
| Path | Effect |
|---|---|
| Behaviour → well-being | β = 0.333 (strongest positive) |
| Stress → well-being | β = −0.256 (strongest negative) |
| Literacy → well-being | β = 0.165 (about half of behaviour) |
| Stress → behaviour | β = −0.353 (stress corrodes habits) |
| Literacy → behaviour | β = −0.027 (not significant) |
That last row is the whole post. The standard policy story is "teach people about money and they'll manage it better." In this data, knowing more did not translate into behaving better. Not a small positive, no significant effect at all.
Stress works through your hands
The second finding is sneakier and maybe more useful. Stressed people behaved worse (β = −0.353), and since behaviour is the biggest lever on well-being, roughly a third of stress's total damage arrived through behaviour (an indirect effect of −0.118). Financial anxiety isn't just unpleasant, it quietly wrecked the habits that the model says matter most. The budget stops getting balanced not because anyone forgot how, but because dread makes the spreadsheet feel like a threat.
Alice reads, Bob automates
Picture two households with the same income. Alice has read every personal finance classic, she can explain marginal tax rates at a barbecue, and every money decision still runs through daily willpower: move the savings manually, decide each month whether to pay the card, feel a spike of dread every time money comes up. Bob knows maybe half of what Alice knows and has set up three boring systems: a paycheque split that moves savings before he sees it, an automatic contribution, and a balanced monthly budget he glances at rather than battles.
The model predicts Bob does better, not because he's smarter (he isn't) but because his outcomes don't depend on how tired, busy or anxious he is this week. Willpower is a terrible co-signer. Automation shows up every month regardless.
Why literacy flopped, and why that's fair
Before the literacy-industrial complex comes for me: this is one study, in one city, in one income stratum, and literacy still showed a positive link to well-being. It's also consistent with the US data on the Big Three quiz, where literacy tracks better outcomes but the mechanism runs through what people do, like planning for retirement, not what they know in the abstract.
Reading about money produces a feeling of progress that requires no action, which is exactly what makes it dangerous. I say that as a person whose product is reading material. The most valuable money content is the kind that ends with you doing one thing this week. Which is my cue.
The honest caveats
One metro area, one country, one income group, so don't transplant the β values onto a Toronto household and call them exact. Cross-sectional surveys can't fully pin down which way the arrows point (stress probably feeds bad behaviour, but miserable budgets also cause stress). Everything is self-reported, and structural equation models on survey data carry standing methodological baggage. The ordering, behaviour above stress above literacy, is the takeaway, not the decimal places.
What to actually do
- Automate one money behaviour this week, before reading another article about money. Contribution, payment, transfer, anything that removes a decision.
- Treat financial stress as a first-class problem to solve, not a personality flaw. It damages outcomes through your behaviour, so it deserves the same seriousness as the budget itself.
- Audit your inputs. If your money content doesn't end in an action, it's entertainment, and you should at least enjoy it honestly as such.
How strong is the evidence?
Scale: Strong = replicated across multiple peer-reviewed studies · Moderate = one solid peer-reviewed study or consistent findings in limited settings · Suggestive = working paper, small samples, or not yet replicated · Thin = non-peer-reviewed source or conflicting evidence.
| Claim in this post | Backing | Strength |
|---|---|---|
| Behaviour predicts well-being about twice as strongly as literacy (β = 0.333 vs 0.165) | Rahman, Isa, Masud, Sarker & Chowdhury — peer-reviewed journal, one cross-sectional survey, 412 households | Moderate — one peer-reviewed study in one limited setting; the post itself warns against transplanting the β values |
| Literacy has no significant effect on behaviour (β = −0.027, n.s.) | same study | Moderate — same single survey; a null in one sample is weak evidence of a true zero, and the broader literacy literature partially conflicts |
| Stress damages well-being partly through behaviour (indirect −0.118) | same study | Moderate — same sample, same caveats; direction consistent with the stress literature |
| Automation beats willpower for money habits | the study's implication, plus adjacent commitment-device research not on this site | Suggestive — the Malaysia study didn't test automation directly; this is inference beyond the paper |
Next steps
- Three questions most people can't answer — the US counterpart: what literacy is worth and where it stops.
- Pay in 4 is a spending button — a habit trap that no amount of literacy defends against.
- The skim view: digest + full PDF — every claim above, one tap from the source.
- The paper itself — Rahman et al, Future Business Journal 7:52, 2021.