By dissecting both business-cycle and long-run fluctuations, this approach strengthens DSGE models for policy analysis.

Read more in my job market paper: sanjaymoorjani.com/uploads/Diss...

Summary: This paper refines our tools for understanding business cycles by:
1️⃣Improved SVAR techniques to identify business cycle shocks
2️⃣Bridging SVAR insights with DSGE estimation

Results align models with data while reducing reliance on questionable assumptions. [16/16]

For instance, backward inflation indexation in DSGE models which is a mechanical way to make the New Keynesian model match the persistence of inflation in the US data is no longer required.

This highlights the value of partial-information estimation over full-information[15/16]

I demonstrate this bias analytically and show its policy implications using the Smets & Wouters (2007) model:
🔹 Estimating the model to match identified business cycle shocks via IRF matching resolves key issues.
🔹 Many ad hoc shocks in DSGE models become unnecessary..[14/16]

I find significant evidence of such non-business cycle long-run shocks.

Separating such non-business-cycle fluctuations is crucial for counterfactual policy analysis, as their inclusion biases parameter estimates of DSGE models. [13/16]

This approach restores the traditional supply-demand dichotomy, modernizing Blanchard & Quah (1989) by identifying a third shock—one driving long-run fluctuations orthogonal to the business cycle—using a novel SVAR methodology...[12/16]

Structural interpretation:
1️⃣ The first shock (no long-run effects) drives positive comovement in inflation & output—similar to demand shocks.
2️⃣ The second shock (both effects) drives negative comovement in inflation & output—consistent with supply or productivity shocks.[11/16]

The two identified shocks explain 99% of business cycle fluctuations of GDP, generating comovements without relying on ex-ante structural restrictions..[10/16]

This paper decomposes macroeconomic fluctuations into three components based on their contributions at different frequencies: one drives business cycles with no long-run effects, another affects both cycles and the long run, and a third impacts only long-run dynamics.[9/16]

Test summarized by the scree plot using the benchmark VAR of ACD, show that business cycle fluctuations are best described as having a two-factor structure: procedures that impose a single source of business cycles risk conflate the effects of two distinct structural shocks[8/16]

This builds on Angeletos et al. (2020) (ACD), who also identify business cycle sources without ex-ante structure. They assume a single “main shock” drives cycles and find “non-inflationary demand shock” as the key driver. But here’s the issue...[7/16]

Key SVAR contribution: I introduce a novel SVAR identification strategy to uncover the causes of business cycles, avoiding any ex-ante assumptions. It can identify drivers like productivity, expectations, financial shocks etc...[6/16]

This approach disciplines DSGE models to focus on relevant moments in the data while reducing the number of parameters. To implement this the paper contributes in two ways to the literature:..[5/16]

1️⃣Use SVAR, which needs fewer assumptions, to identify business cycle shocks.
2️⃣Estimate DSGE models through IRF matching to these identified shocks, avoiding full-information likelihood...[4/16]

Challenge 2: Inclusion of questionable mechanisms (e.g., price indexation).

This results in models with a large number of parameters that are inconsistent with microeconomic evidence.

This paper proposes a two-step solution to such misspecifications in DSGE models: ..[3/16]

Motivation: Chari et al. (2009) argue that New Keynesian models are not useful for quarterly policy analysis. They argue the drive to fit macroeconomic data (likelihood maximization) leads to:

Challenge 1: inclusion of non-structural shocks (e.g., wage markup shocks)..[2/16]

DSGE models are used for the identification of business cycle shocks and serve as a laboratory for counterfactual policy analysis, offering robustness to the Lucas critique. But their reliance on strong microfounded assumptions raises concerns about misspecification..[1/16]