Eve is here. Anyone who has taken a basic economics class has learned that in competitive markets, monopolies and oligopolies command higher prices than sellers, yet Paul, experts, and the press are strangely reluctant to point this out in real-world situations. They seem to have been brainwashed into thinking that bigger = more efficient, and that some of those efficiency savings will be passed on to their customers. The post below is another demonstration that companies in dominant positions can and do dictate prices during periods of rising inflation. This effectively proves the misguided “greedflation” theory.
However, the authors chose not to be so direct. The original article’s headline was “The Detailed Origins of Inflation.” When did “granular” start to be associated with “large”? Read the summary paragraphs below to see how they confound important findings.
Written by Santiago Alvarez-Blazer. Mr. Raphael Auer, Director of BISIH Eurosystem International Payments Center Bank, Sarah Rein, Research Network Fellow CESifo. Faculty member Swiss Institute of Finance. Full professor at Basel University of Macroeconomics. Andrei Levchenko and John W. Sweetland, professors at the University of Michigan Ann Arbor College of International Economics First published: VoxEU
In textbook financial economics, inflation is basically considered to be caused by aggregate shocks such as the money supply and policy interest rates. In this column, I present empirical evidence that inflation is very small-grained and largely influenced by prices set by a small number of large companies. Based on an analysis of 2.9 billion barcode-level transactions across 16 countries, we show that firm-level idiosyncratic shocks account for a significant portion of inflation variation in developed countries. Granular forces are also the main driver of the 2021-2022 inflation shock and have been shown to slow the transmission of monetary policy.
What is the role of large businesses in aggregate inflation, and what are the causes and effects of such “inflation granularity”? Textbook financial economics considers inflation to be primarily caused by aggregate shocks such as money supply and interest rate policy (Woodford 2003, Galí 2015). Although the literature richly models the heterogeneity of micro-level price adjustments, idiosyncratic firm behavior is usually integrated, leaving no role for individual firms in overall inflation. At the same time, following the seminal contribution of Gabaix (2011), an influential macro literature has theoretically modeled and empirically documented that shocks to individual (large) firms can cause aggregate fluctuations.
A recent study (Alvarez-Blaser et al. 2025) shed light on this issue. We analyze a multi-country HomeScan dataset spanning approximately 2.9 billion barcode-level transactions and 16 developed and emerging countries. Each item is linked to a manufacturer, product category, and retailer.
First, we show that the granularity assumption exists in the data because spending is highly concentrated. In the average developed economy, the top 10 companies account for approximately 41% of sales, and the top 10 categories account for approximately 48%. We also see a high degree of synchrony in price changes within multi-product companies and within categories.
We then decompose total inflation into macro components (i.e. weighted averages at the country level) and detailed residuals at the firm and category levels. Our decomposition generalizes traditional granular residual settings (e.g., Gabaix 2011, di Giovanni et al. 2014, Gabaix and Koijen 2024) in two dimensions. First, it allows for multiple non-nested granularity dimensions (company, category, and by extension, retailer). Second, granular residuals may arise from idiosyncratic shocks to large firms or from differences in large firms’ responses to shocks in general. 1 Our concept of granularity residuals explicitly takes into account both of these driving forces. We will document which is more powerful in our context.
Figure 1 shows the high-level results of the analysis. At the macro level, corporate and category disaggregated components account for 56% of the inflation change in developed countries over the period 2005-2020 (see left column of Figure 1). Firm granular residuals are relatively important, explaining about 41% of the variation in inflation. 26 percentage points (dark red) are accounted for by the top 10 companies alone, and the remaining 15 percentage points are accounted for by large companies that are not in the top 10 (light red). The categorical detailed residuals account for an additional 15% of the inflation variance (yellow area). We then decompose the granular residual into components based on their different responsiveness to common and idiosyncratic shocks. The solid granular residue is mainly caused by idiosyncratic shock. In contrast, more than half of the variation in category granularity residuals is due to differences in the responsiveness of categories to common shocks.
Figure 1 Contribution of granularity components to retail inflation fluctuations
Note: This figure shows the percentage of variance in total year-over-year inflation accounted for by each component.
Granularity is less important in emerging markets (see right column of Figure 1). Inflation rates are higher on average and market share is less concentrated. The combined firm and category residuals explain about 20% of the variation in inflation.
Market share concentration and inflation granularity
We then investigate how cross-country differences in inflation granularity are related to market concentration in these countries. In particular, we relate the explanatory power of granular residuals to the market share of top firms. Figure 2 shows a scatter plot of the variance share of total inflation accounted for by the top 10 companies against the average market share of the top 10 companies in each country. There is a statistically significant positive relationship, suggesting that the effect of granularity is stronger in countries with higher market concentration.
Figure 2 Granularity and market concentration
Note: This figure shows a scatterplot of the share of the variance in total year-over-year inflation explained by the firm’s fine-grained residuals relative to the top 10 firms’ share of spending. The dashed line is a linear fit with a slope of 0.78 (robust standard error of 0.31) and R-squared of 0.16 (N=16).
This correlation suggests a trend toward market concentration, as documented, for example, in Autor et al. (2020) may be consistent with the increasing role of firm granularity in aggregate inflation dynamics.
Inflation surge in 2021-22
During the post-pandemic period, the relative importance of granular forces has increased in developed countries. On average, corporate granular factors account for a significant portion of the inflation rate in 2021-2022, reflecting both idiosyncratic shocks and the increased sensitivity of large corporations to common obstacles (supply bottlenecks, energy, etc.). This once again highlights the role played by large companies and price synchronization of multiple products in amplifying macro shocks.
Figure 3 illustrates these findings for Germany. The overall inflation rate (solid line) is shown divided into unweighted average (macro component, dashed line) and granular component (company and sector sum). This figure shows that granular forces played a larger role during the occurrence of inflation.
Figure 3 Germany: Aggregate and detailed components of retail inflation – retailer aspect
Note: This chart shows year-on-year inflation and the sum of each component. Figures for the remaining countries and all available years are provided in the main paper.
Impact on the effectiveness of monetary policy
An important policy consequence is that granularity changes the inflation impulse response to contractionary monetary policy. Local forecast estimates for the United States and the euro area show that aggregate inflation exhibits a short-term “price puzzle,” consistent with previous evidence from Cristiano et al. (1999). This initial increase was driven almost entirely by granular factors, which together accounted for about three-quarters of the short-term rise. In contrast, macro factors did not increase inflation in the year following a monetary policy shock. Consistent with standard theoretical predictions, there was no significant response for about five quarters and then it declined. This suggests that market concentration and the pricing behavior of large firms can slow down disinflation and weaken short-term spillovers. In other words, in a more concentrated environment, financial nonneutrality is formed by the actions of a few large price setters, where financing pass-through and input cost shocks can be more powerful.
Figure 4 Total expansion and component impulse response functions
Note: Panel (a) shows quarterly regional forecast results for all components aggregated into overall year-on-year inflation and macro components only. Panel (b) shows the corresponding results for firm-level and category-level granularity residuals. The shaded area indicates the 90% confidence interval based on the HAC standard error.
conclusion
We show that inflation in developed countries is highly granular, that is, it is heavily influenced by prices set by a small number of large firms. Looking across countries, granularity residuals are less important in economies with less concentrated market shares and higher inflation, such as emerging markets. Disaggregated residuals contributed to the post-COVID-19 inflation spike, with firm-level factors accounting for about a third of inflation in advanced economies in 2021-2022.
Overall, these findings highlight that granular factors beyond traditional macroeconomic determinants are highly important in shaping inflation, which in turn affects the effectiveness of monetary policy and, by extension, its conduct. The operational point is that detailed price microdata for large companies and products is a valuable input for inflation nowcasting. The ECB’s daily price dataset and the BIS Innovation Hub’s Project Spectrum are examples of efforts to develop tools for detailed price-based real-time analysis and nowcasting to improve the extraction of signals from inflation (BIS Innovation Hub 2025).
______
Because we observe prices but not marginal costs, we cannot separate observed price changes into markup adjustments and cost changes. Therefore, our findings do not directly address the recent debate over whether large companies increased their prices disproportionately during the 2021-2022 inflation surge (the so-called “greedflation” or “seller inflation” debate). Available empirical evidence suggests that price adjustments were not the main driver of the inflation spike (e.g., Alvarez-Blaser et al. 2024).
See original post for reference
