Consumption Weighting and Inflation Inequality across Household Groups: Evidence from Malawi

Inflation is often spoken of as a single number, a headline figure that tells a nation how much more expensive life has become. But that single number is an average, a mathematical middle ground that rarely matches the reality of any single family's wallet. When the price of bread rises, the family that buys only bread feels the full shock, while the family that spends most of its money on rent or school fees feels a different, perhaps lighter, pressure. This difference in experience is known as inflation inequality. It matters deeply because governments and aid organizations often use that single average number to adjust wages, set poverty lines, and decide how much money to send to the poor. If the average number does not reflect the specific costs faced by the most vulnerable, those adjustments can fail to protect the people who need them most.

A team of researchers set out to see how this inequality plays out in Malawi, a country in southeastern Africa where many households live on very tight budgets and spend a large share of their money on basic necessities like food and housing. They wanted to know if different groups of people—such as the poor versus the rich, or those living in the countryside versus the city—were actually facing the same rise in prices. To find out, they looked at two distinct periods: a time of moderate price increases between 2016 and 2019, and a much more turbulent period of high inflation from 2019 to 2024, which included the global disruptions of the pandemic and a sharp spike in food costs.

The researchers did not just look at what prices were doing; they looked at what people were buying. They combined detailed records of what thousands of households spent their money on with the official records of how prices for those specific items changed over time. By matching the spending habits of different groups to the price changes they faced, they could calculate exactly how much inflation each group experienced. They focused on six key ways to sort households: by how much money they spent, whether they lived in a city or the countryside, the education level of the head of the household, whether the head was a man or a woman, their age, and the size of the family.

The study revealed a clear and consistent pattern of inequality. In both the moderate and the high-inflation periods, the groups that were already struggling the most faced the highest inflation. Households in the bottom fifth of the spending scale, those living in rural areas, those where the head had not finished primary school, and those headed by a woman all experienced higher price increases than their counterparts. For example, the gap between the poorest and the richest households was significant, with the poorest facing nearly one percentage point higher inflation per year during the moderate period, and that gap widening to almost two percentage points during the high-inflation period. This means that over several years, the cost of living for the poor grew much faster than for the rich, effectively eroding their purchasing power even if their income stayed the same.

The researchers also discovered what was driving these differences. For the poor, rural families, and those with less education, the main culprit was food. Because these households spend such a large portion of their income on food, and because food prices rose sharply, they felt the brunt of the inflation. For female-headed households, the pressure came primarily from housing, water, and energy costs. In contrast, wealthier households and those in cities spent a smaller share of their money on these volatile items and more on things like transport and education, which saw slower price growth. This difference in what people buy is what creates the inequality in inflation exposure.

A crucial part of the study involved how the researchers counted the data. They tested two different ways of averaging the results. One method, which they called democratic, gave every single household an equal voice, regardless of how much money they spent. The other method, which they called plutocratic, gave more weight to the spending of wealthier households, effectively letting their larger budgets dominate the average. They found that while the general pattern of inequality remained the same under both methods, the plutocratic method made the gaps look even larger. This happened because within every group, the richer members tended to have baskets of goods that were less affected by rising prices. When the researchers used the plutocratic method, these richer members pulled the group's average down, making the poorer members' experience seem even more severe by comparison.

The study also uncovered a surprising twist regarding age. When the researchers gave every household an equal voice, younger household heads appeared to face slightly lower inflation than older ones. However, when they weighted the results by how much money was spent, the result flipped, and younger heads appeared to face higher inflation. This shows that the answer to "who is hurt most by inflation" can change depending on whether you are asking about the average person or the average dollar spent. Similarly, the size of the family mattered less in the later period, but the direction of the effect still depended on which counting method was used.

The findings suggest that using a single national inflation number to adjust wages or poverty lines can be misleading, especially in countries like Malawi where poverty is widespread and spending habits vary greatly. The official inflation rate, which is calculated using the plutocratic method and thus reflects the spending of the whole economy, tends to be lower than what the poorest households actually experience. This means that if a government adjusts a poverty line based on the official rate, they might not be giving enough money to keep pace with the actual rising costs faced by the poor. The researchers concluded that to truly understand inflation and its impact on inequality, we must look beyond the single headline number and recognize that different groups live in different price worlds, shaped by what they can afford to buy.