n.V1-3.03| THE TAX ON THINKING
01| The most common explanation for why poor people stay poor is that they make bad decisions. It is not a fringe view and it is not always hostile — it is often held sympathetically, by people who want to fix the decisions. The observation underneath it is not wrong. Sub-optimal choices are made, repeatedly and visibly, and they do make the situation worse. The question is which way the arrow points.
02| Anandi Mani, Sendhil Mullainathan, Eldar Shafir and Jiaying Zhao put the question in the sharpest form available and then tested it:
"The poor often behave in less capable ways, which can further perpetuate poverty. We hypothesize that poverty directly impedes cognitive function."— Mani, Mullainathan, Shafir & Zhao (Poverty Impedes Cognitive Function, Science, Vol. 341, 2013)
03| Their design removed the usual escape route. Indian sugarcane farmers were tested before harvest, when poor, and after harvest, when not poor. Same farmers, same tests, different financial state. Performance was measurably worse in the poor condition. No selection story survives that, because the population is identical on both sides of the comparison. What changed was the pressure, and the pressure took a measurable bite out of the mind doing the deciding.
04| The size is what makes it more than an academic finding. Inducing financial worry costs cognitive performance at magnitudes comparable to a night without sleep. Nobody would ask a person to sit an exam or run an interview immediately after a sleepless night and then treat the result as an estimate of their ability. That is close to what happens every time a person in permanent financial pressure is evaluated on their judgment.
05| The longer version of the same effect shows in development rather than in the moment. Duration of childhood poverty is associated with working-memory performance in adulthood, which suggests the exposure does not simply lift when the pressure lifts. And at population level the relationship runs both ways: poverty raises the incidence of mental illness, and mental illness deepens poverty. Any model with a single arrow in it is wrong at the outset.
06| With that established, the behaviour stops looking like a defect and starts looking like arithmetic. Being poor is expensive in a way that has nothing to do with prices. A missed call because the phone was cut off costs a job. A day in the emergency room over something preventive care would have handled costs an exam. The asymmetry of outcomes is enormous: one mistake and you are done, because there is no buffer standing behind the mistake. Under those conditions the subjective weight of every risk multiplies, and attention compresses toward the next forty-eight hours. Long-horizon planning is a luxury good. It requires a horizon.
07| So the poor are not making bad decisions. They are making defensible decisions under a constraint the observer cannot see, evaluated by an observer who is not carrying it. To someone with a buffer, declining a long-payoff opportunity looks like short-sightedness. To someone without one, accepting it looks like betting the rent on a coin toss.
08| This is where the cause-and-effect inversion does its damage, and it does it in both directions. Those who blame the poor read the behaviour as the origin of the poverty. That is the wrong way round. But the correction is not that the behaviour is harmless. It is real, it is costly, and it obstructs every attempt at a solution — including the sympathetic ones. A programme that requires sustained attention, forward planning and tolerance of delayed reward is asking for exactly the capacities that the conditions of delivery are eroding. That is the tragedy in the structure: the effects of poverty become the obstacles to escaping it, and anyone designing a way out has to solve for both.
09| Which produces a hard constraint on every capability-building proposal in this book part and beyond. Training delivered into conditions of scarcity is working against its own mechanism. Not fatally, and not always — but any proposal that does not say how it handles this has not yet met the strongest finding in the field.
