NBER Data Analysis Finds 'Cheapflation' Causes Higher Inflation for Lower-Income Consumers
A new analysis of billions of retail transactions reveals that official statistics mask how inflation disproportionately impacts budget products, creating a hidden tax on low-income households.
By Sofia Matos
Share this story
In short
- During inflationary periods, the cheapest products experience significantly higher percentage price increases than premium brands.
- Firms tend to pass raw material cost increases to consumers as flat dollar amounts, disproportionately inflating the cost of budget items.
- Because low-income households buy cheaper varieties, they face a substantially higher actual inflation rate than high-income households.
The Bureau of Labor Statistics reports a single inflation number for a broad category like "coffee." But a new wave of economic research analyzing billions of retail scanner transactions reveals a hidden tax on the lowest-income consumers that headline figures completely miss.[1]
The phenomenon is termed "cheapflation." During periods of rising upstream costs, the cheapest products on supermarket shelves experience significantly higher percentage price increases than premium brands sitting right next to them.[2]
A May 2026 working paper published by the National Bureau of Economic Research (NBER) provides the mathematical mechanism behind this disparity. The study analyzed food-at-home sales across more than 30,000 retail stores from 2006 to 2023, tracking exactly how cost shocks ripple down to the consumer.[1]
The core driver is a pricing behavior known as "pass-through in levels." When the cost of a raw input—like agricultural coffee beans—rises, manufacturers and retailers tend to pass that cost onto the consumer on an absolute, dollars-and-cents basis rather than a percentage basis.
The NBER analysis demonstrates how this works in practice. If the wholesale cost of coffee beans increases by 20 cents, a retailer will typically add exactly 20 cents to the price of both a budget coffee brand and a premium artisanal roast.[1]
Because the absolute price increase is identical, the percentage increase is mathematically much larger for the cheaper item. A 20-cent hike on a $2.00 budget coffee represents a 10 percent inflation rate, whereas the exact same 20-cent hike on a $5.00 premium coffee is only a 4 percent increase.[3]
This pricing mechanic directly translates into higher inflation for lower-income households. The NBER data shows that coffee products purchased by the lowest-income quintile are, on average, 24 percent less expensive than those bought by the top quintile.
Consequently, the logarithmic pass-through of cost shocks is substantially higher for cheaper varieties. The lowest unit-price quintile of coffee experienced a 0.57 log point increase, compared to just 0.18 log points for the highest quintile.
This is not an isolated anomaly within a single product category. A separate 2024 NBER study analyzing micro price data from 91 large multi-channel retailers across 10 countries found that regular prices for the cheapest quartile of products grew by an additional 6 to 14 percentage points over premium products during the post-pandemic inflation surge.[2]
By May 2024, the price levels of cheap products had increased by a factor of 1.3 to 1.9 relative to the prices of expensive products, fundamentally altering the purchasing power of budget-conscious shoppers.[2]
Similar patterns emerged in the United Kingdom. Researchers analyzing household scanner data for fast-moving consumer goods found that products on the bottom two rungs of the quality ladder exhibited average price rises of 34 percent between 2021 and 2023, whereas those on the top two rungs rose by only 18 percent.
The reliance on aggregated entry-level item categories by statistical agencies masks this reality. The BLS pools inflation rates across all varieties within a category, effectively averaging out the extremes and presenting a smoothed, middle-of-the-road figure.[1]
According to the 2026 NBER analysis, this aggregation understates the higher cost sensitivity and volatility of food-at-home inflation for low-income households by roughly 70 to 90 percent.[1]
In the disaggregated scanner data, the variance of food-at-home inflation rates is 21 percent higher for the lowest-income quintile than the highest. In the official aggregated data, that gap appears to be just 6.3 percent.
Over the 2021–2023 period, the disaggregated data revealed 2.4 percentage points more food-at-home price growth for the lowest-income quintile relative to the highest. Official income-specific price indices captured only 0.3 percentage points of that gap.[3]
However, the evidence remains bounded by the types of data available. Scanner data excels at tracking fast-moving consumer goods and groceries, but it is less clear if "pass-through in levels" applies equally to services, durable goods, or housing.[3]
Furthermore, researchers note that consumer behavior complicates the data. During inflationary periods, middle-income shoppers often "trade down" to cheaper brands, increasing demand for budget items and potentially giving retailers more pricing power at the lower end of the spectrum.[2]
Definitions
- Cheapflation
- A phenomenon where the prices of lower-tier, budget products rise at a faster percentage rate than premium products within the same category.
- Pass-through in levels
- A pricing strategy where firms pass increased raw material costs to consumers by adding a fixed dollar amount to the price, rather than a percentage markup.
- Scanner data
- Granular, transaction-level data collected directly from retail point-of-sale systems, allowing economists to track the exact prices paid for specific product varieties.
- Logarithmic pass-through
- A mathematical measure of how a percentage change in input costs translates into a percentage change in the final retail price.
Analysis by camp
Distributional Economists
Argue that aggregated inflation metrics hide the true economic burden placed on low-income households.
Researchers focusing on inequality argue that relying on broad entry-level item categories creates a blind spot in macroeconomic policy. By proving that the variance in inflation rates for low-income households is 21 percent higher than for high-income households—compared to the 6.3 percent gap shown in official data—they contend that standard metrics fail to capture lived reality. This camp advocates for integrating granular scanner data into official government statistics to better target fiscal relief during inflationary spikes.
Retail Pricing Strategists
Focus on the mechanical and margin-protecting reasons behind absolute cost pass-through.
From a retail operations perspective, 'pass-through in levels' is not a deliberate attempt to penalize low-income shoppers, but a mechanical necessity. When the wholesale cost of a commodity rises by 20 cents, applying a strict percentage markup to a low-margin budget item might not cover the absolute increase in overhead, packaging, and transport. Retailers apply flat dollar increases to protect baseline margins, which mathematically results in cheapflation.
Macroeconomic Analysts
Emphasize the utility of broad indices for central bank policy despite micro-level flaws.
While acknowledging the distributional disparities revealed by scanner data, macroeconomic analysts argue that central banks like the Federal Reserve must target aggregate demand. Broad indices like the CPI or PCE are designed to measure the overall temperature of the economy, not to serve as a precise cost-of-living index for every demographic slice. They caution that over-complicating headline inflation metrics with micro-level volatility could make monetary policy harder to communicate and execute.
- Distributional Economists
- Argue that aggregated inflation metrics hide the true economic burden placed on low-income households.
- Retail Pricing Strategists
- Focus on the mechanical and margin-protecting reasons behind absolute cost pass-through.
- Macroeconomic Analysts
- Emphasize the utility of broad indices for central bank policy despite micro-level flaws.
Perspectives this story doesn't cover
- Low-income consumer advocacy groups
- Statistical agency methodologists
Sources
Source coverage
3 outlets
3 viewpoints surfaced
- [1]National Bureau of Economic ResearchDistributional Economists
Cheapflation Cycles
Read on National Bureau of Economic Research → - [2]National Bureau of Economic ResearchDistributional Economists
Price Discounts and Cheapflation During the Post-Pandemic Inflation Surge
Read on National Bureau of Economic Research → - [3]Factlen Editorial TeamMacroeconomic Analysts
Synthesis by Factlen editorial team
Read on Factlen Editorial Team →
More in Data & Analysis
See all →Survey Methodology
How Iterative Proportional Fitting Forces Survey Samples to Match Population Margins
5 sources
Statistical Bias
The Mathematics of Attenuation Bias: How Noisy Predictors Flatten Regression Slopes
6 sources
Survey Methodology
Why Complex Survey Designs Lose Statistical Power: Inside the Design Effect Penalty
9 sources
Statistical Theory
Central Limit Theorem: Why Large Sample Averages Form a Normal Distribution
5 sources
Comments
Every angle. Every day.
Get Data & Analysis stories with full source coverage and perspective breakdowns, free every day.