美联储-Modigliani-Miller偏差的综合含义：一种充分的统计学方法（英）-2023.6-76页_609kb

Summary of "Aggregate Implications of Deviations from Modigliani-Miller: A Sufficient Statistics Approach"

This paper introduces a novel methodology—referred to as the sufficient statistics approach—to assess the aggregate macroeconomic implications of firm-level distortions in general equilibrium models without requiring detailed microfoundations. The approach leverages a minimal set of statistics to bound or characterize the effects of deviations from the Modigliani-Miller theorem, such as those due to costly external equity financing and manager-shareholder frictions, on key macroeconomic aggregates.

The core idea is that firm-level distortions affecting investment can be captured by two critical sufficient statistics: the capital-weighted mean distortion to investment rates and the capital-weighted mean-squared distortion to investment rates. These statistics, combined with model parameters, enable the assessment of aggregate outcomes like output, investment, labor, and welfare both in the long run and along transition paths. The methodology proves particularly valuable when full general equilibrium models with microfounded distortions are too complex to solve or when the microfoundations of frictions remain debated.

The authors demonstrate the approach’s applicability by estimating its components using data from corporate finance literature. For extensive external financing frictions (underinvestment), they rely on estimates from Hennessy, Levy, and Whited (2007). For manager-shareholder frictions (overinvestment), they use findings from Ben-David et al. (2013). Calibration of these sufficient statistics and model parameters reveals that aggregate general equilibrium effects are often one order of magnitude smaller than the partial equilibrium (PE) effects previously estimated in friction-specific models. The approach also shows that aggregate inefficiencies (e.g., monopoly markups and corporate taxes) can alter the welfare implications of resolving these frictions.

Key quantitative insights include:

• Removing a friction that causes a 0.13% capital/output ratio distortion leads to approximately a 3.35% increase in output under a partial equilibrium counterfactual.
• In general equilibrium (GE), these effects shrink to around 0.06%–0.09% due to interactions with entry, labor supply, and aggregate price adjustments.
• Misallocation of investment accounts for up to one-third of the aggregate output changes, underscoring the importance of reallocating investment efficiently.
• While resolving a friction always boosts welfare under a friction-removed, revenue-neutral policy when there are no aggregate inefficiencies, inefficiencies can sometimes reduce welfare, necessitating careful consideration of policy interactions.

The paper makes significant contributions by:

1. Introducing a tractable method for assessing aggregate effects of firm-level frictions without detailed microfoundations.
2. Demonstrating the quantitative relevance of these effects, often dwarfing PE benchmarks.
3. Offering extensions to handle richer models (e.g., decreasing returns to scale, endogenous entry, aggregate shocks), ensuring the approach's applicability beyond the baseline framework.
4. Highlighting the interactions between frictions and aggregate inefficiencies, which can significantly alter welfare gains.

This methodology provides a powerful tool for researchers and policymakers to quantify the macroeconomic costs of frictions prevalent in financial and agency settings, bridging the gap between microeconomic evidence and general equilibrium implications.

Key Findings

This paper develops a novel extension to the research by Kurtzman and Zeke (2013) to analyze the aggregate implications of deviations from optimal firm investment rates, while still maintaining insights on the value of the sufficient statistics approach. The paper shows how the approach can be applied to characterize the aggregate effects of firm-level distortions to investment in a broader class of models and provides robust quantitative results across various extensions, including richer firm heterogeneity, deadweight losses, endogenous growth, and time discounting.