The Same Farmer

The month your receivables slip is the month you make the decisions that determine next quarter. There is a body of research suggesting that is not a coincidence, and an equally serious body of criticism arguing about exactly who it applies to.

What This Article Is And Is Not

Stated first, because the subject requires it.

This article describes research on cognitive performance under financial pressure. It is not medical, psychological, financial or legal advice.

Three things it does not do.

It makes no claim about the abilities of any group of people. The central study's strongest design compares the same individuals to themselves at two points in time, which is a claim about a situation rather than about a kind of person, and that distinction is the whole of the article.

It offers no assessment of anyone's cognition, including the reader's.

And it does not address what to do about serious financial distress, which is a matter for a qualified professional and, where relevant, an insolvency practitioner rather than a publication.

If you are under significant financial pressure, the useful move is to speak to an accountant or advisor early, and the research below is part of why.

Our Grades For These Claims

Applying the scheme from the first article in this series.

That financial worry consumes cognitive resources is Grade B. Both the original authors and their critics accept some version of it; they disagree about who it applies to.

The within-person farmer result is Grade B. It is a field study with a strong design, and the authors report it as robust to the specific criticism raised.

That the effect is specific to the poor is Grade C. This is the precise point the published comment contests, and a reanalysis without dichotomising income is reported to fail to corroborate it[3].

Any specific numerical equivalent, such as a stated number of IQ points, is Grade D, and we explain in its own section why we are not using the figures we found.

A Note On Method

Everything here is verified to August 2026.

We obtained the published abstract of the original paper from the publisher, PubMed and a university repository[1][2][6], together with the journal's own editorial summary[2].

We obtained the abstract and opening of the published comment[3] and the full abstract of the authors' response, including a hosted copy[4][5].

We did not obtain the full text of any of the three, and state no effect sizes, no sample sizes and no test scores.

We found a commentary reporting specific test scores that is internally inconsistent, and we set out why we discarded it.

This article reviews behavioural research and is not medical, psychological, financial or legal advice.

The Hypothesis

What was being tested, in the authors' words.

The paper opens: "The poor often behave in less capable ways, which can further perpetuate poverty. We hypothesize that poverty directly impedes cognitive function and present two studies that test this hypothesis."[1]

Two observations, ours, about the logic of that framing.

The first sentence describes an observation everyone already agrees on. What is at issue is its explanation.

And the hypothesis inverts the usual causal direction. The conventional reading of that observation is that some characteristic of a person leads to both poor decisions and poor outcomes. The hypothesis is that the condition itself degrades the decision-making, which makes the observed behaviour a consequence rather than a cause.

That inversion is what the study design is built to test, and it is why the design matters more than any individual number in the paper.

The Mall Study

The first arm.

The authors experimentally induced thoughts about finances and found that this reduces cognitive performance among poor but not in well-off participants[1].

The journal's editorial summary gives the detail. Evidence was gathered from shoppers in a New Jersey mall, and the researchers found that considering a projected financial decision, such as how to pay for a car repair, affects people's performance on unrelated spatial and reasoning tasks[2].

The pattern: lower-income individuals performed poorly if the repairs were expensive but did fine if the cost was low, whereas higher-income individuals performed well in both conditions, as if the projected financial burden imposed no cognitive pressure[2].

Three observations, ours.

The tasks were unrelated to money. Spatial and reasoning problems, not budgeting. So whatever happened was not a matter of financial skill.

Nothing had actually happened to anyone. The manipulation was a hypothetical scenario, which makes the effect, if real, remarkably cheap to trigger.

And this arm is the one the published criticism targets, for reasons set out below. It is the weaker half of the paper.

The Farmers

The second arm, and the stronger one.

The authors examined the cognitive function of farmers over the planting cycle and found that the same farmer shows diminished cognitive performance before harvest, when poor, as compared with after harvest, when rich[1].

The journal's summary confirms the direction plainly: the sugarcane farmers from Tamil Nadu performed these tasks better after harvest than before[2].

Two observations, ours.

Sugarcane farming supplies something a researcher can almost never arrange: a large, predictable, exogenous change in a person's finances, arriving on a schedule set by agriculture rather than by anything about the farmer.

And the same people were measured on both sides of it. That is the feature the next section is about, and it is what makes this the part of the paper worth reading.

Why The Design Is The Point

The methodological core. This section is our own analysis.

Almost every study of poverty and cognition is between-person: it compares poorer people to richer people. That design is close to uninterpretable, because income correlates with education, childhood nutrition, health, schooling quality, language exposure and a dozen other things that also affect performance on a reasoning test.

The farmer study is within-person. It compares the same man to himself, months apart.

Which means every stable characteristic is held constant by construction rather than by statistical adjustment. Genetics, education, childhood nutrition, language, schooling, personality and innate ability are all identical between the two measurements, because it is the same person.

Three consequences.

The design eliminates the entire class of confounds that make between-person comparisons on this topic worthless, and it does so without needing to model them.

It makes the finding a claim about a situation, not about a category of person. Nothing about the farmer changed. His circumstances did.

And that is not merely an ethical nicety, it is the logical content of the result. A within-person effect cannot support a statement about kinds of people, because no kind of person varied.

What The Authors Ruled Out

The obvious alternative explanations, and their fate.

The paper states the finding cannot be explained by differences in time available, nutrition, or work effort[1].

And separately: "Nor can it be explained with stress: Although farmers do show more stress before harvest, that does not account for diminished cognitive performance."[1]

Three observations, ours.

Those are the four explanations a sceptical reader would reach for first, and the authors addressed them in the abstract, which is where an author puts a defence they expect to need.

The stress result is the interesting one. Farmers were more stressed before harvest. The stress simply did not account for the cognitive difference, which distinguishes this from a general claim that pressure makes people worse at thinking.

And we did not obtain the analyses supporting any of these exclusions, and report them only as claims made in the abstract.

The Proposed Mechanism

The explanation offered.

The authors conclude that poverty itself reduces cognitive capacity, and suggest this is because poverty-related concerns consume mental resources, leaving less for other tasks[2].

The journal's summary phrases it as poverty imposing a cognitive load that saps attention and reduces effort[2].

Two observations, ours.

The model is capacity, not capability. Nothing is claimed to be permanently reduced. The proposal is that a fixed resource is being spent elsewhere, in the way a computer running a large background process is slower at everything without being a worse computer.

Which implies reversibility, and that is what the harvest data is meant to show. The same farmer scores higher once the financial situation resolves.

The Published Criticism

The formal objection, published in the same journal.

Wicherts and Scholten published a comment in Science, 342(6163), 1169[3].

Their argument: "A reanalysis without dichotomization of income fails to corroborate their findings and highlights spurious interactions between income and experimental manipulation due to ceiling effects caused by short and easy tests."[3]

Two technical points there, and both are serious. Explained plainly, and this explanation is ours.

Dichotomisation means splitting a continuous variable, here income, into two groups. Doing so discards information and is known to be capable of producing effects that are not present in the underlying continuous data. The critics reanalysed without splitting and report the findings not holding up.

Ceiling effects mean a test is too easy, so scores bunch at the top. If well-off participants are near the maximum in both conditions, they cannot show a decline, and an apparent interaction between income and condition can be an artefact of the test rather than a fact about the people.

What The Critics Actually Concluded

The part of the criticism that is almost never reported, and it changes the practical implication entirely.

Wicherts and Scholten's stated conclusion: "This suggests that effects of financial worries are not limited to the poor."[3]

Three observations, ours, and this is the most important passage in the article.

The critics are not arguing that financial worry has no cognitive effect. They are arguing that the interaction is spurious, meaning the effect is not specific to low-income participants.

Which means the two camps agree on more than the dispute suggests. Both accept that financial worry consumes cognitive resources. They disagree about whether being well-off protects you from it.

And for a business owner, the criticism is the more concerning of the two positions. The original paper implies a comfortable reader is insulated. The criticism implies nobody is. A reader who dismisses this literature because it was contested has dismissed it in the wrong direction.

The Authors' Response

The reply, published alongside.

Mani, Mullainathan, Shafir and Zhao responded in Science, 342(6163), 1169e[4].

They address the criticism in three numbered parts: "(i) using a continuous income variable, the interaction between income, and experimental manipulation remains reliable across our experiments; (ii) our results in the cognitive control task do not appear driven by ceiling effects; and (iii) our observed post-harvest improvement is robust to the presence of learning."[4][5]

Three observations, ours.

The response is point-for-point. Continuous income answers dichotomisation, ceiling effects are addressed directly, and the third point answers a criticism about the field study.

Point (iii) is worth naming because the objection is obvious once stated: farmers took the same tests twice, so they might simply have got better at the tests. The authors report the improvement is robust to that.

And the hosted response contains the observation that the relevant data are, as both parties recognise, noisy[5]. That is a candid thing for an author to concede in a defence.

A Third Line Of Criticism

A separate objection, aimed at the mechanism rather than the statistics.

A commentary published in Frontiers in Psychology questions the legitimacy of using the limited-resource model to interpret the possible association between poverty and self-regulation failure, and investigates the impact of poverty induced distraction on tasks relying on proceduralized processes[7].

We did not obtain this commentary beyond its indexed description and report nothing further about it.

Two observations, ours.

This attacks a different layer. The statistical comment disputes whether the effect is where the authors say it is. This disputes whether the bandwidth explanation is the right account of it.

And the limited-resource framing has form. The first article in this series described ego depletion, an extremely influential limited-resource model of self-control, collapsing under a 23-laboratory replication and bias correction. Anyone reading a resource metaphor in psychology should hold it loosely, and that includes this one.

Numbers We Declined To Use

A note on something we found and discarded.

We located secondary commentary reporting specific Raven's test scores for the farmers, giving 5.45 and 4.35 as the pre-harvest and post-harvest averages respectively, and stating in the same sentence that this indicated a notable cognitive performance improvement after income stabilization.

Those two statements are mutually inconsistent. If performance improved after harvest, the post-harvest figure cannot be the lower one. The journal's own summary confirms the farmers performed better after harvest than before[2], so the ordering in that commentary is wrong or the labels are reversed.

We also found the widely circulated claim that the effect equates to roughly thirteen IQ points. That figure does not appear in any abstract or journal summary we obtained, and reaches us only through secondary commentary.

We are therefore reporting no test scores and no IQ equivalence anywhere in this article.

Two observations, ours.

The IQ-point figure is the single most quoted thing about this study, and we could not verify it. Anyone using it should be quoting the paper.

And this is the sixth article in this series to find a famous study's headline numbers reported unreliably by secondary sources. The pattern is now consistent enough that the secondary number should be treated as unverified by default.

Where We Land

Our own reading, stated so a reader can disagree with it.

Four propositions.

Financial worry consumes cognitive resources. Both sides of the published dispute accept a version of this.

The within-person farmer result is the strongest evidence here, because its design eliminates the confounds that make this question hard, and the authors report it robust to the learning objection.

Whether the effect is specific to the poor is genuinely unresolved, and it is the precise point in dispute.

And the bandwidth mechanism is a hypothesis rather than a finding, contested separately, and belonging to a family of limited-resource models that has not fared well.

Our summary: the phenomenon is better supported than its explanation, and its boundary conditions are less settled than either camp's headline suggests.

The Reading This Does Not Support

Stated explicitly, because the topic invites the opposite conclusion. This section is our own.

Nothing in this literature supports the view that poorer people are less capable.

Three reasons, and the first is decisive.

The strongest result is within-person. The same farmer, months apart. No comparison between people is being made at all in that arm, so it cannot license a conclusion about people.

The authors' framing is explicitly the reverse. They open by noting the poor often behave in less capable ways and hypothesise that this is caused by the condition rather than being a characteristic that produces it[1].

And the critics push further in the same direction, concluding effects of financial worries are not limited to the poor[3].

Both camps, then, describe a situation people move in and out of. The dispute is over its boundaries, not over whether it is a property of persons.

The Owner Under Cash Pressure

The application to this publication's reader. This is our own extension and was not studied.

A business owner in a cash squeeze has the structural features this research describes.

A pressing financial problem that will not resolve on its own, occupying attention outside the moments allocated to it.

Consequential unrelated decisions arriving in the same period: hiring, pricing, whether to take a contract, whether to renew a lease.

And no external event that resolves it on a schedule, unlike a harvest.

Three cautions, firmly.

The studies concern mall shoppers and subsistence farmers, not business owners, and an owner with an overdraft is not in the position of a farmer before harvest.

If the original authors are right that the effect concentrates among the poor, it may not apply to a solvent business owner at all.

If the critics are right that it is not limited to the poor, it applies to everyone, which is the reading under which this section matters.

We cannot tell you which. We can tell you the question is unresolved in the published literature and that both answers are live.

Tunnelling And Overborrowing

A related finding by overlapping authors, reported briefly.

An indexed description of Shah, Mullainathan and Shafir (2012), Some Consequences of Having Too Little, Science 338, 682–685, states that scarcity changes how people allocate attention: it leads them to engage more deeply in some problems while neglecting others, and that scarcity leads to attentional shifts that can help to explain behaviors such as overborrowing[8].

We did not obtain that paper and report only this description.

Two observations, ours.

This is a different claim from the bandwidth one. It is not that capacity falls but that attention narrows, with deeper engagement on the pressing problem and neglect elsewhere. Those are distinguishable, and a business could exhibit either.

And overborrowing is named as an example. For a business under cash pressure, that is the specific decision most likely to be taken during the squeeze and most likely to extend it.

The Loop

The structure both literatures point at. Ours, and offered as reasoning.

Financial pressure occupies attention. Attention is what consequential decisions require. Consequential decisions taken under pressure are worse. Worse decisions extend the pressure.

Two observations.

If that loop is real, then the cost of a cash squeeze is not only the squeeze. It is the decisions taken during it, whose consequences arrive later and are usually attributed to something else.

And it implies the timing of a decision is a variable you control in a way the pressure often is not. A pricing decision or a lease renewal does not always have to be settled this month.

What Slack Is For

The constructive implication, and it is a financial one. Ours.

Every business text argues for a cash buffer on the grounds that it absorbs shocks. This literature suggests a second reason.

Three points.

A buffer removes the financial question from the foreground, which is the resource this research says is being consumed.

Its value is therefore partly in decisions never observed: the bad contract not signed, the panic discount not offered. Which means, as in six previous articles in this series, the benefit leaves no trace and will not appear in any measurement.

And that asymmetry is an argument for holding a larger buffer than a purely liquidity-based calculation would justify, because the liquidity calculation prices only the shocks and not the judgment.

We present that as reasoning from a contested literature, not as a finding, and certainly not as financial advice.

What To Do

Do not read this as being about kinds of people. The strongest result compares the same farmer to himself, which cannot support a claim about categories of person.

Note that the criticism widens the finding rather than killing it. The published comment concludes effects of financial worries are not limited to the poor.

Treat the timing of consequential decisions as a variable. If a pricing or hiring decision can wait until after the receivable lands, that costs nothing to arrange.

Be most careful about borrowing during a squeeze. A related paper names overborrowing as a behaviour scarcity helps explain, and it is the decision most likely to extend the pressure.

Get the financial question out of the foreground. A resolved or scheduled problem occupies less than an open one, even at the same amount owed.

Count the buffer's second benefit. A liquidity calculation prices the shocks a buffer absorbs and not the decisions it prevents, and the second leaves no trace.

Speak to an advisor early rather than late. If the loop is real, the period when you most need outside judgment is the period when your own is most occupied.

Distrust the thirteen IQ points figure. It is the most quoted claim about this study and we could not verify it in any abstract or journal summary.

The Limits Of This Analysis

Several caveats matter. This article reviews behavioural research and is not medical, psychological, financial or legal advice, and makes no claim about the abilities of any group of people. Anyone under significant financial pressure should speak to a qualified professional. Everything is verified to August 2026. We did not obtain the full text of the original paper, the published comment, or the authors' response, relying on abstracts, the journal's editorial summary and a hosted copy of the response. We state no effect sizes, no sample sizes and no test scores. We report the authors' exclusion of time, nutrition, work effort and stress as claims made in the abstract, having obtained none of the supporting analyses. We found and discarded secondary figures that are internally inconsistent, and we could not verify the widely quoted thirteen IQ point equivalence in any abstract or journal summary. We did not obtain the Frontiers commentary beyond its indexed description, nor the 2012 paper on attentional shifts, and report each only as described. Whether the effect is specific to lower-income participants is the precise point in published dispute and is unresolved. The bandwidth mechanism is a hypothesis rather than a finding and is separately contested. The studies concern mall shoppers and subsistence farmers; every application to business owners in this article is our own extension to a population that was not studied, and may not hold. The loop, the argument about slack, and the reading of what the criticism implies are our own reasoning, not findings.

Related Reading

Behavioural Finance Not A Medical Condition The other case where a condition of the environment is misread as a property of the person. No. 18 What Your Own Business Is Worth Owners want more than buyers will pay. A famous experiment said that was loss aversion, a second said it was an artefact, and a third failed to find it... Behavioural Finance Modestly Better Another study whose most quoted number could not be traced to the paper.

Frequently Asked Questions

What did the study actually find?

That the same farmer showed diminished cognitive performance before harvest, when short of money, compared with after harvest. The authors report this cannot be explained by time available, nutrition, work effort, or stress, and propose that financial concerns consume mental resources.

Is this saying poorer people are less capable?

No, and the design rules that reading out. The strongest result compares the same individuals to themselves months apart, so no comparison between people is being made. It is a claim about a situation people move in and out of, and the authors' own framing is that the condition causes the behaviour rather than the reverse.

Was the study criticised?

Yes, formally, in the same journal. The comment argued that a reanalysis without dichotomising income failed to corroborate the findings and that apparent interactions arose from ceiling effects caused by short and easy tests. The authors responded point by point, and both are published.

So should I dismiss the finding?

Read what the critics concluded: that effects of financial worries are not limited to the poor. They are not arguing the effect is absent, but that it is not specific to low-income people. For a business owner that is the more concerning position, not the reassuring one.

What about the thirteen IQ points figure?

We could not verify it. It does not appear in any abstract or journal summary we obtained and reaches us only through secondary commentary, so we report no IQ equivalence and no test scores at all. We also found and discarded secondary scores that contradicted themselves.

What is the practical takeaway for a business?

Treat the timing of consequential decisions as something you control, be especially careful about borrowing during a squeeze, and speak to an advisor early. If the loop is real, the period when you most need outside judgment is the period when your own is most occupied.

References

1. Mani, A., Mullainathan, S., Shafir, E., & Zhao, J. (2013). Poverty impedes cognitive function. Science, 341(6149), 976–980. DOI 10.1126/science.1238041, published abstract via PubMed, on the poor often behaving in less capable ways which can further perpetuate poverty; on the hypothesis that poverty directly impedes cognitive function; on the authors experimentally inducing thoughts about finances and finding this reduces cognitive performance among poor but not well-off participants; on the same farmer showing diminished cognitive performance before harvest, when poor, compared with after harvest, when rich; on this not being explicable by differences in time available, nutrition, or work effort; and on it not being explicable by stress, since although farmers show more stress before harvest that does not account for the diminished cognitive performance. Note: we obtained the published abstract only; no effect sizes, sample sizes or test scores are stated anywhere in this article. pubmed.ncbi.nlm.nih.gov
2. Publisher record and editorial summary for Mani and colleagues (2013), Science, on lacking money or time leading to poorer decisions, possibly because poverty imposes a cognitive load that saps attention and reduces effort; on evidence being gathered from shoppers in a New Jersey mall and farmers in Tamil Nadu, India; on considering a projected financial decision, such as how to pay for a car repair, affecting performance on unrelated spatial and reasoning tasks; on lower-income individuals performing poorly if repairs were expensive but fine if the cost was low, whereas higher-income individuals performed well in both conditions as if the projected financial burden imposed no cognitive pressure; on the sugarcane farmers performing these tasks better after harvest than before; and on the suggestion that poverty-related concerns consume mental resources, leaving less for other tasks. Note: the journal's own record and editorial summary. science.org
3. Wicherts, J. M., & Scholten, A. Z. (2013). Comment on "Poverty Impedes Cognitive Function". Science, 342(6163), 1169. DOI 10.1126/science.1246680, on Mani and colleagues having presented laboratory experiments aiming to show that poverty-related worries impede cognitive functioning; on a reanalysis without dichotomization of income failing to corroborate their findings and highlighting spurious interactions between income and experimental manipulation due to ceiling effects caused by short and easy tests; on this suggesting that effects of financial worries are not limited to the poor; and on the authors criticising the results on statistical and psychometric grounds. Note: a peer-reviewed published technical comment; we obtained the abstract and opening only. science.org
4. Mani, A., Mullainathan, S., Shafir, E., & Zhao, J. (2013). Response to comment on "Poverty impedes cognitive function". Science, 342(6163), 1169e. DOI 10.1126/science.1246799, institutional repository record reproducing the abstract, on Wicherts and Scholten having criticised the study on statistical and psychometric grounds; and on the authors showing that using a continuous income variable the interaction between income and experimental manipulation remains reliable across their experiments, that their results in the cognitive control task do not appear driven by ceiling effects, and that their observed post-harvest improvement is robust to the presence of learning. Note: a university repository record of a peer-reviewed published response. collaborate.princeton.edu
5. Author-hosted copy of the response to the comment, reproducing the three numbered rebuttal points, recording the reference to Wicherts and Scholten at Science 342, 1169 (2013), and noting that the relevant data, as both Wicherts and Scholten recognise and the authors point out, are noisy. Note: an author-hosted PDF; we obtained a portion. zhaolab.psych.ubc.ca
6. University repository record for Mani and colleagues (2013), confirming authorship, journal, volume, issue and pages, and reproducing the abstract identically. Note: an institutional repository record, used as an independent confirmation of the citation and abstract text. collaborate.princeton.edu
7. Indexed record for a commentary in Frontiers in Psychology on Poverty impedes cognitive function and The poor's poor mental power, questioning the legitimacy of using the limited-resource model to interpret the possible association between poverty and self-regulation failure, and investigating the impact of poverty induced distraction on tasks relying on proceduralized processes; and recording the accompanying Perspective, Vohs, K. D. (2013), The poor's poor mental power, Science, 341(6149), 969–970. Note: we obtained the indexed description only and report nothing further about this commentary's contents. ncbi.nlm.nih.gov
8. Semantic Scholar record for Mani and colleagues (2013) with indexed descriptions of related work, on the suggestion that poverty-related concerns consume mental resources leaving less for other tasks; and on Shah, A. K., Mullainathan, S., & Shafir, E. (2012), Some Consequences of Having Too Little, Science, 338, 682–685, suggesting that scarcity changes how people allocate attention, leading them to engage more deeply in some problems while neglecting others, and that scarcity leads to attentional shifts that can help explain behaviours such as overborrowing. Note: a bibliographic record with indexed descriptions. We did not obtain the 2012 paper. semanticscholar.org

This article reviews behavioural research and is not medical, psychological, financial or legal advice. It makes no claim about the abilities of any group of people. Anyone under significant financial pressure should speak to a qualified professional. No paper discussed was obtained in full; no effect sizes, sample sizes or test scores are reported. The widely quoted IQ-point equivalence could not be verified and is not used. Whether the effect is specific to lower-income participants remains in published dispute. Every application to business owners is the authors' own extension to a population that was not studied.