A Look at Inflation in Recent Years through the Lens of a Macroeconomic Model

In recent years, the U.S. inflation rate reached levels not seen since the early 1980s. Year-over-year personal consumption expenditures (PCE) inflation—the Federal Reserve’s preferred measure—peaked at 7.2% in June 2022, while core PCE inflation—another inflation measure closely watched by policymakers—hit 5.6% in February 2022. This surge in inflation was closely linked to the significant economic fluctuations around the onset of the COVID-19 pandemic. However, economists remain divided on its precise causes, citing factors such as pent-up consumer demand, global supply chain disruptions, and expansionary fiscal and monetary policies during this period.

For instance, former Fed Chair Ben Bernanke and Olivier Blanchard argued that inflation was primarily driven by external price shocks, highlighting volatility in global commodity prices and supply chain disruptions. They attributed the rise in inflation to “supply factors,” which are external cost pressures passed on to consumers through higher prices.

However, Domenico Giannone and Giorgio Primiceri argued instead that “demand forces” were the main drivers of elevated inflation during this period. These demand forces are changes in the components of domestic expenditure, such as consumption, investment or government spending, which induce firms to raise production to keep up with demand. These increases in production often result in higher marginal costs of production, as firms need to pay higher wages to attract more workers, for example, thus leading them to eventually raise prices.

To the extent that expansionary fiscal policy involves increases in government consumption or investment or increases in transfers to households and businesses that boost private consumption and investment, it is considered one such demand factor. Expansionary monetary policy, too, is often classified as a demand factor, as the traditional monetary policy transmission mechanism postulates that keeping interest rates lower helps stimulate private consumption and investment for a given price level.

In this blog post, I use an estimated macroeconomic model of the U.S. economy to analyze the recent inflationary trends and decompose them into the aforementioned factors. The model’s results align more closely with Giannone and Primiceri’s perspective, suggesting that demand factors—particularly expansionary fiscal and monetary policies—played a central role in driving inflation after the onset of the pandemic. Importantly, these same factors also contributed to a robust economic recovery during this period.

The Macroeconomic Model

The analysis is conducted using a version of the St. Louis Fed DSGE model, a medium-scale dynamic stochastic general equilibrium (DSGE) model of the U.S. economy. DSGE models are mathematical representations of the economy that simulate how various agents interact and respond to different shocks. These agents are:

• Households, which consume, save and invest
• Firms, which produce goods and services, hire workers, and rent capital
• The government, which collects taxes, spends and provides transfers
• The central bank, which sets interest rates to influence monetary policy

These agents make dynamic decisions, accounting for how today’s actions affect their future welfare. Their interactions across multiple markets determine the general equilibrium of the economy. The model is “stochastic,” meaning that it accounts for unpredictable shocks—such as changes in productivity or preferences—that influence agents’ decisions.

To accurately reflect the U.S. economy, key parameters are calibrated to long-term averages or estimated using historical data from the first quarter of 1959 to the fourth quarter of 2019. This approach allows the model to perform tasks ranging from forecasting to counterfactual policy analysis. Crucially, the model can interpret data through a historical shock decomposition, identifying the external shocks and policies that best explain observed economic behavior.

“Supply and Demand” vs. “Structural Shocks”

The debate on the causes of inflation is often framed as a contest between supply and demand. This line of thinking draws on basic economic principles:

• “Demand shocks” are interpreted as being shocks that move the rates of inflation and economic growth in the same direction, just like an exogenous shift in the demand curve for a given product would move both the quantity and price of that product in the same direction.
• “Supply shocks,” on the other hand, are those that move the rates of inflation and output growth in opposite directions.

Adam Shapiro, for example, applies this logic on several PCE categories to study the relative contributions of demand and supply to the recent behavior of inflation.

In contrast, the DSGE model adopts a structural approach. Aggregate supply and demand dynamics emerge as outcomes of agents’ microeconomic behavior and market interactions. Inflation, therefore, results from the interplay of numerous structural shocks—such as changes in productivity growth, labor market disruptions or variations in household savings preferences. While these structural shocks can be grouped into “demand” (e.g., fiscal and monetary policies, foreign demand) and “supply” (e.g., productivity growth, labor market fluctuations) categories for simplicity, such classifications are somewhat arbitrary.

Understanding Inflation through the Lens of the Model

The first figure decomposes year-over-year inflation into the components that are driven by structural shocks in the demand group versus the supply group. We focus on core PCE inflation, which excludes energy and food prices and is therefore the measure of inflation that the model is best suited to capture. (The last observation for the fourth quarter of 2024 is derived from the core PCE nowcast published by the Cleveland Fed.) This type of exercise is called a historical shock decomposition, and it allows researchers and policymakers to use an estimated model in order to study the contribution of different shocks to movements in a variable that exists both in the data and within the model.

The plot presents the historical shock decomposition in terms of percentage point deviations of core PCE inflation from 2%, so the zero line corresponds to 2% inflation. The figure reveals that while supply factors were relevant in the early stages of the COVID-19 period, they were more than offset by negative demand factors, which resulted in inflation consistently below 2% in 2020. Starting in 2021, demand factors began to exert upward pressure on inflation, while supply factors subsided. The subsequent decline in inflation starting in 2022 coincides with a decline in these demand pressures, with the recent uptick at the end of 2024 being explained by resurgent supply-side inflationary pressures. In conclusion, the decomposition points to “demand”-related factors as having been relatively more important in explaining both the rise and decline of inflation in recent years.

Year-over-Year Core PCE Inflation Broken Down by Type of Structural Shock

SOURCES: Bureau of Economic Analysis and author’s calculations.

NOTE: Fourth-quarter data for 2024 are derived from the core PCE nowcast published by the Cleveland Fed.

Identifying the Impact of Fiscal and Monetary Policies

We can go one step further and decompose the fiscal and monetary policy components of these demand shocks. That exercise is presented in the next figure, which separates fiscal and monetary policy contributions to core PCE inflation. Fiscal policy contributes to inflation via changes in government spending and transfers to households, while monetary policy contributes to inflation by deviating from a monetary policy rule estimated on historical data (a variant of the Taylor rule) and setting interest rates at a lower level than such a rule would warrant.

The analysis suggests that fiscal policy played an important role in raising the inflation rate between 2020 and 2022, while monetary policy contributed to inflation between 2021 and 2023. Throughout 2023, the role of monetary policy diminished, but the effect of demand remained elevated due to other demand-related factors (such as households’ increased desire to consume for a given price level and interest rate).

Year-over-Year Core PCE Inflation Broken Down by Key Sources

SOURCES: Bureau of Economic Analysis and author’s calculations.

NOTES: Fourth-quarter data for 2024 are derived from the core PCE nowcast published by the Cleveland Fed. The percentage points of certain sources, such as “other,” may be too small to be visible.

Conclusion

Historical shock decompositions of core PCE inflation using the St. Louis Fed DSGE model reveal that structural shocks classified under the “demand” category played an important role in both the rise and subsequent decline of U.S. inflation in recent years. The contribution of policy, both monetary and fiscal, was particularly important.

Yet these same fiscal and monetary policies also appear to have supported the U.S. economy while it underwent shocks during the pandemic. The next figure plots a historical shock decomposition of the U.S. output gap, which is the difference between actual output (as measured by gross domestic product) and the level of output that would prevail in the absence of a series of frictions. (Fourth-quarter data for 2024 are estimates calculated by the St. Louis Fed DSGE model.)

Output Gap Broken Down by Key Sources

SOURCES: Bureau of Economic Analysis, Federal Reserve Bank of Atlanta and author’s calculations.

NOTES: Fourth-quarter data for 2024 are estimates calculated by the St. Louis Fed DSEG model. A positive output gap indicates the economy is producing above its potential capacity, measured by production in a counterfactual economy without nominal rigidities, while a negative output gap indicates the economy is producing below its potential capacity.

The output gap is often interpreted as a measure of economic slack. As shown in the figure above, the output gap analysis suggests that macroeconomic policies not only fueled inflation but also supported economic recovery by countering severe negative shocks in 2020 and 2021.In the second quarter of 2020, the Federal Reserve aggressively eased its monetary policy. But under the output estimate derived from the St. Louis Fed DSGE model and the Taylor rule variant, monetary policy had a negative effect on the output gap in that quarter due to institutional constraints, such as the zero lower bound, as well as other factors such as endogenous persistence generated by frictions and other features of the model. This underscores the traditional stabilizing role of policy during crises.

Note

1. In the second quarter of 2020, the Federal Reserve aggressively eased its monetary policy. But under the output estimate derived from the St. Louis Fed DSGE model and the Taylor rule variant, monetary policy had a negative effect on the output gap in that quarter due to institutional constraints, such as the zero lower bound, as well as other factors such as endogenous persistence generated by frictions and other features of the model.