Monetary Policy News and Long-Run Inflation Expectations: Measuring News Where Expectations Are Formed
Hiroki Yamamoto (Bank of Japan); with Masato Nishiuchi
Session 1A · Fri 20 Nov, 10:45–11:15 · Room A

Can monetary policy move long-run inflation expectations? We identify a perceived monetary policy shock from professional forecasters’ same-target forecast revisions and compare it with a realized-side shock identified under the same restrictions. When entered separately, ten-year inflation expectations respond to both shocks. When entered jointly, however, the response to a perceived monetary tightening remains persistently negative, while the realized shock adds essentially no incremental explanatory power. This response survives comparisons with high-frequency and narrative monetary policy measures and remains in quarters in which the federal funds policy setting is unchanged. Whether monetary policy reaches long-run inflation expectations is therefore empirically inseparable from where policy news is measured.

Learning, Heuristics and Anchored Inflation: How do Different Types of Consumers Change Their Minds about Inflation?
Kalvinder Shields (University of Melbourne); with Kevin Lee, Viet Nguyen
Session 1A · Fri 20 Nov, 11:15–11:45 · Room A

Bivariate VAR models of actual and expected inflation are estimated using alternative estimation methods to investigate how, in real time and as information arrives, different types of consumers learn about inflation. The estimated models distinguish various views about learning including those based on Bayesian updating or on heuristics. Using a detailed Australian survey, we find that learning is best characterised by periodic shifts in beliefs rather than incremental adjustments, that different consumer groups learn at different times and from different experiences, and that their long-run inflation expectations occasionally become unanchored and take a range of values which are inter-connected but do not move in sync.

Wages when Inflation is Lower or Higher than Expected
Nalini Prasad (UNSW Sydney)
Session 1A · Fri 20 Nov, 11:45–12:15 · Room A

This paper explores how wage-setting behaviour adjusts to changes in the inflation environment. I utilise data from 185,000 multi-period wage agreements between firms and it workers in Australia. I find that negotiated wage growth responds asymmetrically to inflation surprises: negative surprises have little effect while positive surprises raise wage growth. However the pass-through of inflationsurprises is nonlinear; while moderate positive inflation surprises lead to higher wage growth, the marginal response declines sharply for large shocks. During the post-pandemic high-inflation period, negotiated wage growth did not rise proportionally with inflation surprises. I show that these findings can be rationalised by a model of state-dependent wage setting in which workers respond to unexpected real wage losses while firms face convex costs of nominal adjustment.

Job Ladder with Career Requirements
Yota Isobe (Kyoto University)
Session 1B · Fri 20 Nov, 10:45–11:15 · Room B

This paper simultaneously explains the labor share decline and the rise in wage inequality using an on-the-job search model. In the model, firms are heterogeneous in productivity. When firms post vacancies, they can specify qualifications for applicants, based on the applicant’s current job. This feature endogenously determines workers’ possible transition path from current to new employers. I then introduce affine transformations of productivity and unemployment payoff levels. I first derive analytical results for job-specific wage shares, the distribution of workers across firms, and the Gini coefficient. In the results, the two parameters of the affine transformation play distinct roles. The difference stems from the fact that proportional change raises poaching incentives and thus job-to-job mobility, whereas a common level shift leaves these incentives unchanged. The model calibrated to U.S. data shows that the multiplicative parameter can lower the labor share and raise wage inequality, whereas the additive parameter tends to generate the opposite outcomes.

Lifelong Gains from a Strong Start: Early Career Fit and the College Wage Premium
Paul Jackson (National University of Singapore); with Ismail Baydur, Jie Duan
Session 1B · Fri 20 Nov, 11:15–11:45 · Room B

This paper studies the role of career uncertainty in shaping the college wage premium. We document stark differences in the first half of workers’ careers by educational attainment. Non-college workers separate into unemployment and switch careers frequently while exhibiting sluggish wage growth. College graduates separate and switch careers at a lower rate, experience rapid wage growth, and form more accurate expectations about their future occupation. We build a life cycle search model in which workers who are more certain they are well matched to their career invest more in human capital accumulation and see faster wage growth. Equalizing career uncertainty across education groups raises non-college wages by 14.3% and reduces the college wage premium by 17.9%. Non-college workers who fail to find their best career fit early in the baseline economy but do so in the counterfactual enjoy lifelong gains, earning 29% more over their working lives.

Mid-Worklife Crisis
Nicolas Jacquet (Singapore Management University); with Paul Jackson, Serene Tan
Session 1B · Fri 20 Nov, 11:45–12:15 · Room B

This paper studies the aggregate labor-market consequences of mid-worklife crisis in a heterogeneous-agent life-cycle macroeconomics model. We document two new U.S. facts: (i) unemployment rate among college (non-college) graduates is U-shaped (declining monotonically) over the life-cycle; and (ii) the U-shape for college graduates has become more pronounced over time. We develop and calibrate a life-cycle directed-search model for college graduates using CPS and NLSY with one novel feature - employed workers can gain/lose skills – and replicate life-cycle patterns. Older unemployed workers are less attractive to firms because they have fewer years left to work, are costlier to hire, and have smaller upside. Combined, these three effects generate rising unemployment rates from middle-age on and their strengthening in recent times account for the increasingly pronounced U-shape. Our model predicts that 10% of young workers with high human capital face substantial downward mobility; and raising retirement age from 60 to 70 reduces old workers’ unemployment rate by 10%.

Decomposing Loan Rate Dispersion in the Interbank Market
Masayuki Okada (Hitotsubashi University)
Session 1C · Fri 20 Nov, 10:45–11:15 · Room C

This paper decomposes loan rate dispersion in the interbank market. Using confidential transaction-level data, we document dispersion in uncollateralized overnight loan rates even during periods of low interest rates in Japan. Applying the AKM model, we quantify the contributions of borrower and lender heterogeneity to loan rate dispersion. Lender heterogeneity explains about three times more dispersion than borrower heterogeneity. The contribution of borrower heterogeneity declines over time. Positive assortative matching in borrowers’ and lenders’ average loan rates is observed. For term loans, borrower heterogeneity accounts for a larger share of loan rate dispersion than for overnight loans.

Heterogeneous Lenders in Credit Markets (HELIX)
Yasin Onder (Ghent University); with Juan Carlos Hatchondo
Session 1C · Fri 20 Nov, 11:15–11:45 · Room C

Sovereign bonds are held by investors who differ markedly in their willingness to carry default risk across time, yet quantitative models of sovereign default almost universally price debt with a single representative lender. This paper develops a quantitative model of sovereign borrowing, default, and renegotiation in which long-duration bonds are traded by two types of risk-averse lenders—patient and impatient—who differ in their discount factors. Because there is no representative pricing kernel, the bond price cannot be read off a single Euler equation: it emerges as a fixed point at which each type’s portfolio choice satisfies its own optimality conditions and both the defaultable and risk-free asset markets clear. Debt restructurings take the form of a stochastic bargaining game, and we study how collective action clauses (CACs) interact with the distribution of bond holdings across lender types: which group commands the supermajority required to approve an exchange shapes both the incidence of settlement and recovery rates. The framework delivers time variation in the identity of the marginal pricer of sovereign risk, endogenous migration of debt toward patient lenders in bad times, and a channel through which the composition of the investor base affects borrowing costs, default incentives, and the design of restructuring proposals.

Information Acquisition and the Finance-Uncertainty Trap
Ding Dong (Hong Kong Baptist University); with Allen Hu, Zhaorui Li, Zheng Liu
Session 1C · Fri 20 Nov, 11:45–12:15 · Room C

This paper studies the interplay between firms’ information acquisition and credit constraints, leveraging novel measures of firm-level information acquisition constructed from online job postings data. Exploiting exogenous variation in information supply by financial news media, we show that shocks raising the cost of information acquisition tighten firms’ credit access. Complementing this evidence, credit tightening is associated with persistent declines in information acquisition, particularly for firms in industries more dependent on external finance. To examine the macroeconomic consequences of this feedback loop between information acquisition and credit access, we develop a tractable general equilibrium model featuring heterogeneous firms, endogenous signal precision, and value-based borrowing constraints. The model characterizes a ``finance-uncertainty trap’’: high information costs elevate uncertainty and depress firm equity, hampering credit access; in turn, tighter credit reduces the scale of production and the returns to learning, further discouraging information acquisition. Calibrated to match micro-level moments, the model shows that this feedback loop substantially amplifies and prolongs business cycle fluctuations. Patterns from oil supply news shocks are consistent with the model’s prediction that news-driven declines in expected equity value and borrowing capacity are associated with lower information acquisition.

High-Dimensional Nonparametric Local Projections
Jiaming Huang (Hong Kong University of Science and Technology); with Òscar Jordà, Massimiliano Marcellino, Tommaso Tornese
Session 1D · Fri 20 Nov, 10:45–11:15 · Room D

This paper develops a two-step high-dimensional nonparametric local projections estimator (HDNLP) for average structural functions and nonlinear impulse responses with high-dimensional controls. The first step constructs Neyman-orthogonal pseudo-outcomes with high-dimensional nuisance functions, and bypasses cross-fitting under neighborhood-stability conditions. The second step regresses the pseudo-outcome on low-dimensional policy variables using a series approximation, which delivers pointwise and uniform inference for potentially nonlinear responses. Under explicit high-level conditions, we establish a uniform Gaussian approximation, feasible confidence bands, and specification tests for parametric response shapes, and we derive lower-level rate bounds with a set of primitive conditions on the data and nuisance function learners.

Macroeconomic Forecasting with subset regularization
Bonsoo Koo (Monash University); with Didier Nibbering
Session 1D · Fri 20 Nov, 11:15–11:45 · Room D

Macroeconomic forecasting typically combines a small set of core variables with many auxiliary indicators whose predictive strength is uncertain. We propose subset regularization, which allows strong predictors to enter unregularized, shrinks weak predictors to balance bias and variance, and excludes noise variables. We show that subset regularization improves forecast accuracy relative to the common macroeconomic practice of preselecting predictors to enter unregularized. The method jointly selects this allocation and the regularization parameter by minimizing recursive forecast loss, using a multi-start stochastic search for practical computation. We establish conditions under which the algorithm attains the oracle population MSFE target. For predictor sets larger than the sample, a frozen-screening extension yields an MSFE oracle inequality and approximate recovery of the screened oracle allocation. An empirical application to forecasting U.S. inflation and industrial production shows that the common approach of preselecting unregularized predictors in macroeconomic forecasting models is outperformed by subset regularization.

Exact Identification of Demand and Supply Shocks
Raul Santaeulalia-Llopis (New York University Abu Dhabi); with Andre Groeger, Luis Rojas, Yanos Zylberberg
Session 1D · Fri 20 Nov, 11:45–12:15 · Room D

Macroeconomics is organized around aggregate demand and aggregate supply, and hence around the identification of demand and supply shocks. Existing empirical strategies recover these objects under distinct—and typically mutually exclusive—restrictions on shock properties and transmission (e.g., timing, sign, or orthogonality), yielding contrasting policy prescriptions from the same data. We provide a novel theoretical framework for the exact identification of aggregate demand and supply shocks. By leveraging joint information on the subjective expectations and realizations of prices, quantities, and factor inputs, our approach achieves identification while remaining fully agnostic on shock properties and transmission dynamics. We apply this framework to a unique firm-level panel from the Bank of Italy (INVIND) to construct new empirical shock series.

Price Rigidity or Cost Dormancy? Firm-Level Evidence on the Phillips Curve
Yojiro Ito (Bank of Japan); with Shunsuke Haba
Session 2A · Fri 20 Nov, 13:30–14:00 · Room A

The output-based Phillips curve combines a pricing-side response to marginal cost with the elasticity of marginal cost to output. Using revenue and expenditure data for 35,000 Japanese firms over 2000–2022, we recover firm-level prices and marginal costs and estimate annual static cost-price pass-through together with the components of the marginal-cost elasticity. Pass-through is large and rises over time. The wage-plus-curvature component of the cost elasticity is small and shows no trend; the wage contribution is small, and the estimated input-price response to industry output is not statistically distinguishable from zero at the five percent level. Wage-bill and employment elasticities to firm-level output roughly halve over twenty-one years. The evidence points to a weak activity-cost link despite high annual cost-price pass-through.

Unpacking Global Inflation
Benjamin Wong (Monash University); with Yun Young Gwak, James Morley
Session 2A · Fri 20 Nov, 14:00–14:30 · Room A

We reinterpret global inflation by separating its imported component from commonality due to broader nominal or demand forces, such as similar policy responses. To do so, we construct a quarterly international CPI dataset that disaggregates headline inflation into goods and services components for 42 countries over 1971–2023 and estimate a sectoral dynamic block factor model with global commodity prices. Three findings emerge. First, sectoral disaggregation reveals multiple sources of global inflation that matter at different horizons: imported shocks dominate in the short run, while broader nominal or demand forces remain relevant at longer horizons. Second, global shocks account for a large share of goods and headline inflation variation and a smaller but meaningful share of services inflation variation. Third, after 2001, the global component of headline and goods inflation increases mainly at short horizons and shifts toward imported channels. These results imply that global inflation should not be equated with imported inflation: although imported shocks have become more important in the short run, broader common nominal and demand forces remain an important part of global inflation and are more closely connected to the margins affected by domestic policy.

Diminishing Monetary Potency under Fiscal Dominance: A Bayesian Local Projections Analysis
James Morley (University of Sydney); with Naveed Javed
Session 2A · Fri 20 Nov, 14:30–15:00 · Room A

We consider empirically how fiscal stance influences the potency of monetary policy. New Zealand provides a compelling laboratory to study this form of monetary-fiscal interaction given its stable history of inflation targeting along with substantial changes in its fiscal stance due to global forces acting on a small open economy (SOE). A simulation-based Bayesian Local Projection (BLP) framework is developed to estimate the macroeconomic effects of a narrative measure of monetary policy shocks when there are many possible omitted variables, especially in this SOE setting. Our BLP approach incorporates a novel shape prior on impulse response functions to help manage the substantial bias-efficiency tradeoffs given the relatively small effective sample sizes when considering nonlinearities inherent in policy interactions. For a smooth-transition regime-switching model with an endogenously-estimated threshold parameter, we find that monetary policy is clearly more potent, especially in terms of output and inflation, when there is a high degree of fiscal consolidation. Consistent with the fiscal theory of the price level, our results for a more general model that also allows for sign asymmetries suggest the effects of expansionary versus contractionary monetary policy shocks on output, inflation, and the exchange rate are actually reasonably symmetric, implying that the potency of monetary policy is more related to monetary-fiscal dominance than coordination.

Labor Market Dualism as an Equilibrium
Ji-Woong Moon (Chung-Ang University); with Serena Rhee
Session 2B · Fri 20 Nov, 13:30–14:00 · Room B

This paper studies the origins and efficiency implications of a dual labor market, characterized by the coexistence of permanent and fixed-term contracts. We show that dualism can arise even under optimally chosen labor market institutions. To do so, we build a directed search model with incomplete contracts, in which firms cannot commit to productivity-contingent wages and firing decisions, leading to excessive separations. The employment protection mitigate these excessive firings, but protection is costly to firms and is paid for with fewer jobs. A fixed end date trades job security for job creation on a different margin than the protection does. We show that when matches are inspection goods, there exist workers for whom the vacancy creation induced by a shorter horizon outweighs the security it forgoes, even under an optimally chosen level of protection. These workers prefer a temporary contract, and the dual labor market emerges as part of the optimal design. We further characterize equilibrium sorting across contract types and the type-specific optimal duration of temporary contracts, and draw implications for observed policies such as tenure-graded protection and probation periods. We calibrate the model to South Korea to characterize the optimal employment protection and contract duration.

Beyond Flexibility: Pitfalls in Liberalizing Temporary Workers
Mitsuru Katagiri (Waseda University)
Session 2B · Fri 20 Nov, 14:00–14:30 · Room B

This study examines the effects of liberalizing the use of temporary workers within a general equilibrium model with heterogeneous firms. Given stringent employment protection legislation (EPL) for regular workers, liberalizing temporary workers leads to more efficient labor allocation across firms while deterring human capital accumulation among workers. A simple analytical model shows that EPL induces firms to rely excessively on temporary employment because firms internalize severance payments as private costs even when they are transfers from a social perspective. A quantitative analysis using Japanese data implies that liberalizing temporary workers could deteriorate social welfare and that easing EPL for regular workers significantly improves social welfare not only by realizing more efficient labor allocation across firms but also by reducing the share of temporary workers.

Technical Change, Task Allocation, and Labor Unions
Martyna Marczak (Trinity College Dublin); with Thomas Beissinger, Franziska Brall
Session 2B · Fri 20 Nov, 14:30–15:00 · Room B

We integrate task-based production with labor union wage setting and search-and-matching frictions in a general equilibrium framework. We uncover a previously unrecognized mechanism: task reallocation, here in the context of skill-biased technical change (SBTC), affects union wage markups by altering the wage elasticity of labor demand. This mechanism depends on the shape of the relative task productivity schedule that describes comparative advantage of high-skilled workers in more complex tasks. If SBTC-induced task reallocation raises wage pressure sufficiently strongly, low-skilled workers suffer from higher unemployment or lower wages. We calibrate the model using German and French data for 1995–2005 and 2010–2017. Simulations show that SBTC increases low-skilled unemployment in both countries in the first period, and in Germany also in the second one. This adverse impact is alleviated by shorter contract durations, which reduce wage pressure. We also provide empirical evidence for the shape of the relative task productivity schedule to validate the calibration.

Time-varying Effective Lower Bound Tightness
Hibiki Ichiue (Keio University)
Session 2C · Fri 20 Nov, 13:30–14:00 · Room C

The sensitivity of long-term yields to macroeconomic conditions reflects the severity of the effective lower bound (ELB) on the short-term policy rate or the remaining monetary policy space. Using individual professional forecasts for Japan, we estimate monthly expected sensitivities of the Japanese Government Bond (JGB) yield to inflation and real activity as an inverse indicator of ELB tightness. The estimates suggest that the ELB tends to be more binding when the long-term yield is lower, inflation is lower, or market stress is higher. However, the ELB has no clear relationship with the short-term interest rate, and evidence for a correlation with the long-term yield is relatively weak. Moreover, some portion of the fluctuation remains unexplained by these macroeconomic and financial factors, likely reflecting shifting perceptions of the Bank of Japan’s reaction function.

Asymmetric interest rate smoothing and financial stability
Joseph Kachovec (Reserve Bank of Australia); with Isaac Green, Scott Abrahams
Session 2C · Fri 20 Nov, 14:00–14:30 · Room C

Central banks mechanically lower policy rates at a faster pace than they raise them. Using a panel dataset of policy rates from 38 countries, this paper shows that during easing cycles, central banks reach terminal rates more quickly than during tightening cycles. This asymmetry persists after controlling for inflation and output gaps in Taylor-type rules, suggesting it reflects a policy preference rather than responses to economic fundamentals. We find that this asymmetry is more pronounced in countries with higher financial stability risks as measured by bank capital ratios, net interest margins, and credit-to-GDP ratios. Our results suggest an implicit interaction between monetary policy and financial stability mandates that is consistent with policy concerns about macro-financial linkages.

Shortages, Liquidity, and the Impact of Monetary Policy on Prices: Theory and Evidence from Commodity Markets
Yuki Sato (The University of Tokyo)
Session 2C · Fri 20 Nov, 14:30–15:00 · Room C

We find that monetary policy shocks have no significant effect on commodity futures prices when markets are in backwardation, that is, when futures prices lie below spot prices. Because backwardation signals supply constraints, reflecting a positive convenience yield that arises under tight supply or low inventories, this finding reveals a new channel for understanding the effects of monetary policy under supply shortages. Our model to investigate this mechanism relies on the inventory arbitrage of commercial traders and the rational inattention of financial traders, which together replicate the higher price response to the aggregate demand shock under ample inventories than under tight inventories. To reinforce our empirical findings, we conduct two complementary analyses. The first tests the assumption that the basis is informative about storage conditions by constructing an Inventory Constraint Index, which closely tracks observed inventory data from the World Agricultural Supply and Demand Estimates and the U.S. Energy Information Administration. The second examines the liquidity channel, an alternative mechanism through which monetary policy may affect commodity futures markets. Both analyses support our interpretation: while illiquidity is priced through changes in interest rates, inventory tightness, as indicated by the basis, remains the primary transmission channel.

Monetary-Fiscal Nexus and Unintended Equilibria
Daisuke Ikeda (Bank of Japan)
Session 2C · Fri 20 Nov, 15:00–15:30 · Room C

Multiple equilibria challenge rule-based central banks with inflation targets. This paper shows that, with non-Ricardian households, multiple steady states (SSs) may prevail in a positive interest environment. While a targeted SS always exists, an additional SS, characterized by higher or lower inflation, emerges depending on monetary-fiscal interactions. If the stabilization stance of monetary or fiscal policy is weak, a high-inflation SS arises with determinacy, while the targeted SS loses determinacy, featuring indeterminacy or explosiveness. A low-inflation SS is also prone to indeterminacy or explosiveness. The results imply potential risk of high inflation in an economy with non-Ricardian households.

Using GPU Parallelization to Solve Heterogeneous-Agent Models with Aggregate Shocks
Robert Kirkby (Victoria University of Wellington)
Session 2D · Fri 20 Nov, 13:30–14:00 · Room D

We adapt the Matched-Expectations Path algorithm for the global, nonlinear solution of heterogeneous-agent models with aggregate shocks to the GPU. The key adaptations are: (i) after matching, the value function can be solved in parallel across time periods; (ii) exploiting conditional monotonicity of the policy function reduces both memory use and runtime; and (iii) adding a second-layer interpolation grid makes interpolation parallelizable. These algorithms are implemented in VFI Toolkit, and are thus easily available to users. Examples include a standard heterogeneous-agent real business cycle (HA-RBC) model, uncertainty shocks in an HA-RBC model, entrepreneurial choice in an HA-RBC model, and quasi-hyperbolic discounting in an HA-RBC model.

Monotone Mixing and Distribution Dynamics
John Stachurski (National Graduate Institute for Policy Studies); with Qingyin Ma, Takashi Kamihigashi
Session 2D · Fri 20 Nov, 14:00–14:30 · Room D

Many economic applications establish stability of Markov dynamics using the monotone mixing condition (MMC) of Hopenhayn and Prescott (1992). Working in the same setting as that paper, we introduce a weak monotone mixing condition (WMMC), phrased in terms of the reversal of rank between populations, and show that it is strictly weaker than the MMC and both necessary and sufficient for global stability under the Kolmogorov metric. We apply these results to a small open economy version of the Aiyagari model, showing how the WMMC can be used to establish global stability in a setting where the MMC fails. In addition, we show that, when the state space is one-dimensional, the MMC and WMMC coincide, implying that the original MMC is necessary as well as sufficient in this setting.

Faithful Decoding
Jingni Yang (University of Sydney); with Nisha Peng, John Stachurski, Ziyue (Humphrey) Yang
Session 2D · Fri 20 Nov, 14:30–15:00 · Room D

This paper studies transformations that increase efficiency in solving equilibrium systems without information loss. Our approach exploits order-theoretic structure commonly found in economic problems to obtain conditions under which high-dimensional systems can be transformed into low-dimensional systems while preserving exact relationships between their solutions. The transformations can also be used for purposes other than dimensionality reduction, such as simplifying analysis and facilitating stochastic approximation routines. The theoretical ideas are illustrated using applications from economics and finance. In a real option problem, we demonstrate speed gains of up to 70,000 times.

Price versus Quantity: Dual-Track Transmission of Chinese Monetary Policy
Dora Xia (ASEAN+3 Macroeconomic Research Office); with Tom Chenyang Wang
Session 3A · Sat 21 Nov, 09:30–10:00 · Room A

This paper proposes a new approach to measuring the stance of Chinese monetary policy by separately identifying its price-based and quantity-based tools. The price-tool indicator measures common movements in interest rates administered by the People’s Bank of China (PBC), while the quantity-tool indicator captures the central bank’s aggregate liquidity injections and withdrawals. We then use a Bayesian VAR with zero and sign restrictions to examine how these shocks propagate through different economic channels. Quantity-based tightening produces a rapid and substantial contraction in real activity through an immediate decline in bank lending. By contrast, price-based tightening affects market interest rates more directly but is transmitted to the real economy with longer lags and a more muted impact. Our results suggest that the two sets of tools affect the real economy through different transmission channels, with quantity-based tools remaining more potent for now.

Assessing the Impact of Trade Policy Uncertainty on Asset Pricing: A Convenience Yield Channel
Pratiti Chatterjee (University of Western Australia)
Session 3A · Sat 21 Nov, 10:00–10:30 · Room A

Macro-finance studies typically examine U.S. dollar convenience yields in episodes of financial distress, where the flight-to-safety channel dominates. This article reassesses the global role of the U.S. dollar through the lens of trade policy uncertainty (TPU). Using an SVAR on monthly data from 2000 to 2025, a TPU shock generates real dollar depreciation, declines in industrial production and interest rates, tighter financial conditions, and widening convenience yields. To explain these patterns, I build a two-country New Keynesian model with financial intermediaries subject to balance-sheet constraints. Heightened TPU expands intermediary risk, driving up dollar convenience yields. At the same time, precautionary motives depress demand and trigger monetary easing, reconciling lower bond yields and exchange rate depreciation with widening liquidity premia. The findings suggest that although the shock is contractionary and induces real currency depreciation, the U.S. dollar’s dominance in the broader financial system remains resilient.

Prices or Paychecks: Measuring Macroeconomic Preferences with Conjoint Analysis
Özer Karagedikli (Asia School of Business); with Andrew Coleman, Sui-Jade Ho
Session 3A · Sat 21 Nov, 10:30–11:00 · Room A

This paper uses survey-based conjoint analysis to estimate the trade-offs a representative sample of Americans are willing to make to achieve desired inflation, unemployment, income, and exchange rate outcomes. The estimates suggest people would on average be indifferent between a 1 percentage point increase in the inflation rate, a 1.75 percentage point increase in unemployment, a 1.33 percentage point decrease in income growth, or an 8.33 percent depreciation of the exchange rate. The average masks substantial heterogeneity that is only weakly related to standard demographic and socioeconomic characteristics. Rather, people can be clustered into four distinctive preference groups that possibly reflect latent beliefs, narratives, or subjective models of the economy. Utilizing the Federal Reserve’s ‘Tealbook’ forecasts from 2013 and 2019, we use the estimates to rank alternative monetary policy options. In 2013, most households would have considered the baseline monetary policy path a good compromise between tighter and looser alternatives. In contrast, in 2019 there would have been substantial opposition to flexible price-level targeting and substantial support for the tighter monetary policy associated with a Taylor rule.

Unbundling Labor
Chris Edmond; with Simon Mongey
Session 3B · Sat 21 Nov, 09:30–10:00 · Room B

Why is there so much wage inequality within occupations? Why has the overall level of within-occupation wage inequality risen? Why is the rise concentrated in cognitive-intensive occupations, while inequality has fallen in manual-intensive ones? To answer these questions we develop a general equilibrium model with heterogeneous workers who supply multiple skills jointly, as an indivisible bundle. This bundling friction allows a given skill to be priced differently in different occupations. We find in US data that differences in skill prices, not worker composition explain why within-occupation inequality is higher in more cognitively intensive occupations and account for about a third of the rise in within-occupation inequality from 1980 to 2010. We show that a uniform increase in the productivity of cognitive skill drives inequality up in cognitive-intensive occupations and down in manual ones, just as in the data.

Why Do Americans No Longer Work So Much More Than Non-Americans?
Kurt See (National University of Singapore); with Serdar Birinci, Loukas Karabarbounis
Session 3B · Sat 21 Nov, 10:00–10:30 · Room B

In the 1990s, Americans used to work much more than non-Americans. Nowadays, about half of the gap in hours worked has reversed. To evaluate the convergence of working hours, we develop a tractable model of labor supply enriched with multiple sources of heterogeneity across individuals, an extensive margin of participation, multi-member households, and an elaborate system of taxes and benefits upon non-employment. Using detailed measurements from micro-level and aggregate datasets, we identify model parameters and sources of heterogeneity across individuals for various countries. We run a horse race between competing explanations and find that U.S. hours per person declined after 2000 owing mainly to the rise of government health benefits provided to the non-employed. Non-U.S. countries have generous benefits for the non-employed, but this generosity has not changed as much over time as in the United States, and public health coverage does not depend on employment status or income levels. For these countries, the rise of labor supply is generally accounted for by a mix of factors, such as the rise of wages and the falling disutility of work.

Dual Income Earners and Productivity
Cristiano Mantovani (University of Melbourne); with Shu Lin Wee
Session 3B · Sat 21 Nov, 10:30–11:00 · Room B

We examine when joint households contribute towards productivity improvements. Motivated by the increasing share of females in joint households employed in full-time non-routine cognitive jobs, we develop a directed search model with heterogeneous job choice. Income pooling and consumption security afforded by one’s partner affect the decision of whether to accept a risky job that grows one’s productivity. Using our estimated model, we find that females in joint households contributed the most to productivity improvements. Declining non-employment risk was a significant factor driving these gains, while changes in the marginal distributions of underlying types played a smaller role.

Cash Transfers in Developing Countries: Local Multipliers, Aggregate Effects, and External Adjustment
Wataru Miyamoto (University of Hong Kong); with Arthur Mendez, Thuy Nguyen Lan, Steven Pennings
Session 3C · Sat 21 Nov, 09:30–10:00 · Room C

What are the macroeconomic impacts of cash transfers in developing countries, and how do they change as programs scale up? To answer these questions, we build a two-region New Keynesian small open economy model and calibrate it to rich microdata from a randomized controlled trial (RCT) in Malawi. The model implies a local multiplier on temporary transfers of about one in treated regions, which depends on the degree of price stickiness, the share of liquidity-constrained households, and the extent of local bias in consumption. Spillovers to the rest of the country can be negative if the capital account is closed, as the transfer can appreciate real exchange rate and crowd out exports, but can be positive when capital can flow more freely and even larger when the exchange rate is fixed. A counterfactual scaled-up cash transfer program is likely to have smaller local and aggregate multipliers in Malawi, mostly due to its closed capital account. Our analyses suggest that macroeconomic frictions and the monetary policy regime are crucial to pin down the aggregate effects of scaled-up programs from local RCT estimates.

The Dynamics of Agricultural Production: Technological Constraints, Misallocation, and Crop-Specific Price Shocks
Luis Rojas (New York University Abu Dhabi); with Alaïs Martin-Baillon, Raul Santaeulalia-Llopis
Session 3C · Sat 21 Nov, 10:00–10:30 · Room C

Agricultural productivity in rural economies is not only low but also highly dispersed across farms. We argue that both patterns partly reflect sluggish production adjustments to increasingly volatile agricultural conditions. To investigate this mechanism, we develop and estimate a dynamic model of agricultural production in which technological constraints (time to grow) and market frictions limit farmers’ ability to adapt to changing conditions. We discipline the model using the adjustment of production—factors, inputs, and cropping patterns—to large shocks affecting the returns to different crops across 220 Vietnamese villages. We find that: (i) staggered responses to shocks account for a substantial share of land productivity dispersion across farms, and (ii) both technological and market frictions matter quantitatively, with the latter generating significant misallocation in land use and cropping patterns.

Exchange Rate Transmission through Multinational Firms: Evidence from Japan
Hyunseung Oh (Federal Reserve Board); with Ryan Kim, Bin Ni, Choongryul Yang
Session 3C · Sat 21 Nov, 10:30–11:00 · Room C

We study an income-account channel of exchange-rate transmission using matched data on Japanese multinational parents and their foreign affiliates. During the sharp yen depreciation of 2021–22, we find that (i) foreign affiliates with greater exchange-rate exposure expanded their activity and generated more profits, with larger payments to their Japanese parents, and (ii) more exposed parents increased their average wages and shifted employment toward commercial branches and offices involved in sales, purchasing, and related business functions, with offsetting declines in other employment. A multinational firm model in which scarce parent-side support inputs are allocated between domestic operations and foreign affiliates rationalizes these findings. To study the aggregate implications, we embed this mechanism in a two-country general-equilibrium model with trade and multinational production. The model shows that moving foreign-market activity from exports to affiliate production reduces the domestic real GDP response to depreciation and shifts more of the external adjustment from trade and toward income earned abroad. Depreciation therefore need not generate a large expansion in domestic production.

Estimating Heterogeneous Economies with Micro Data
Tommy Iao (Reserve Bank of Australia); with Yad Selvakumar
Session 3D · Sat 21 Nov, 09:30–10:00 · Room D

We give sufficient conditions under which dynamic equilibrium models with heterogeneous-agents can be represented by a first-order reduced-rank vector autoregression. We exploit this result to develop an econometric framework that enables the rapid estimation of a rich class of models with macro and repeated cross-sections of micro data. Simulation evidences suggest that our method delivers increased precision of parameter estimates than conventional approaches, particularly for parameters governing cross-sectional dynamics. We apply our method to estimate a medium-scale HANK model with heterogeneous earnings exposures to aggregate fluctuations at the household-level. Our estimates imply that low-income households are more sensitive to changes in aggregate income on average, and that this sensitivity is heightened conditional on monetary policy shocks. Through the lens of the model, our method estimates that heterogeneous earnings exposures amplify the aggregate consumption response to monetary policy shocks by 40%, substantially larger than those implied by traditional estimation methods

Precautionary savings motive in a volatile economy: Evidence from Argentine Administrative Data
Emiliano Carlevaro (Adelaide University); with Juan Zurita
Session 3D · Sat 21 Nov, 10:00–10:30 · Room D

The precautionary saving hypothesis implies that consumers save to protect themselves against labour income volatility. Evidence on this mechanism in emerging economies is scant and is often limited to aggregate cross-country data. We study the effect of labour income uncertainty on saving using a long panel of individual-level salary data from Argentina. We extend a standard model of individual labour income to accommodate multiple sectors of the economy. Using this model and the salary data, we estimate labour income volatility across sectors. We then estimate its effect on savings using district-level data on bank deposits and a spatial dynamic panel model. We find that labour income volatility has a modest positive effect on savings. The spatial model indicates that most of this effect arises from spatial spillovers from nearby districts.

Tax Avoidance and Wealth Inequality
Jose Alvarado (University of Queensland)
Session 3D · Sat 21 Nov, 10:30–11:00 · Room D

The hypothesis that wealth inequality is driven by the higher returns earned by the rich—thereby offsetting the progressivity of the tax system—overlooks a key dimension: tax avoidance. This paper shows that tax avoidance not only undermines the progressivity of the tax system but is also one of the reasons why the wealthy earn higher returns. Using micro-data from Chilean tax records, I quantify tax avoidance and find that the top 0.01% of taxpayers reduce their tax payments by 80% through corporate investments. To measure the impact of tax avoidance on wealth inequality, I calibrate a Bewley-Huggett-Aiyagari heterogeneous agent model, incorporating two departures from standard approaches: (i) endogenous portfolio choices between safe and corporate risky assets, and (ii) tax functions that account for tax avoidance. The model successfully replicates the 50% wealth share held by the top 1% in Chile. The main intuition is that, given the presence of tax avoidance, the after-tax rate of return on risky assets increases, leading agents to reallocate their portfolios towards these assets, ultimately resulting in an even higher rate of return on wealth. The main quantitative result is that, without tax avoidance, the top 1% wealth share decreases from 50% to 11%. These findings suggest that tax avoidance is a key driver of wealth inequality.

Monetary Policy and Diagnostic Asset Price Bubbles
Ernie Chuang (National Taiwan University); with Been-Lon Chen
Session 4A · Sat 21 Nov, 13:40–14:00 · Room A

Diagnostic expectations provide a realistic behavioral model of inference. This paper revisits Galí (2014) by replacing rational expectations with diagnostic expectations in a model with asset- price bubbles. We show that when agents strongly extrapolate and bubble shocks are persistent, a policy of “leaning against the wind”—raising interest rates in response to rising asset prices—can stabilize bubbles, reversing Galí’s result. Our mechanism differs from Miao et al. (2019), which relies on targeting an alternative steady state, and Allen et al. (2025), which emphasizes sufficiently high interest rates to shrink bubbles contemporaneously.

Asset Price Dispersion, Monetary Policy and Macroprudential Regulation
Simon Mishricky (Australian National University)
Session 4A · Sat 21 Nov, 14:00–14:20 · Room A

Following the 2007–2009 Global Financial Crisis, sustained monetary expansion and tighter financial regulation have left financial markets thinner, less resilient, and more prone to instability. This paper develops a monetary model of decentralised financial exchange to account for these outcomes. The framework links search frictions and costly posting to the joint effects of monetary and regulatory policy on asset prices, quoting behaviour, and market stability. Increases in inflation or posting costs reduce quoting intensity, widen the distribution of executable prices, and raise the probability of trading breakdowns. The model replicates key post-crisis patterns such as wider spreads, higher execution costs, and an increased likelihood of flash-crash events, showing that the interaction between monetary and regulatory policy can unintentionally increase financial market fragility.

Collateralized Borrowing as Double-Edged Sword: Asset Encumbrance and Bank Fragility
Mingyang Liu (The University of Tokyo)
Session 4A · Sat 21 Nov, 14:20–14:40 · Room A

This paper studies how collateralized borrowing affects bank fragility in a dynamic general equilibrium model with bank runs. Motivated by the 2022–2023 U.S. banking crisis, I extend Gertler and Kiyotaki (2015) by introducing collateralized borrowing as a third funding instrument alongside deposits and net worth. Collateralized borrowing serves as a double-edged sword: it relaxes the banker’s incentive constraint and expands funding capacity, but encumbers assets in a bank run because secured creditors hold seniority over depositors. Calibrated to the 2022–2023 crisis episode, the model shows that collateralized borrowing cushions bank balance sheets against adverse shocks by allowing bankers to sustain capital holdings about 13% higher than in a deposits-only benchmark, yet raises the peak run probability by nearly 50%. A counterfactual decomposition shows that asset encumbrance accounts for roughly 60% of this risk increase, dominating the over-leveraging channel. Finally, a policy experiment shows that the Bank Term Funding Program, by valuing collateral at face value rather than market value, reduces asset encumbrance and cuts the peak run probability by about 75%.

Auditing Decisions and Income Tax Evasion
Md Khademul Islam Chowdhury (Australian National University); with Jiu Lian, Shahar Rotberg
Session 4B · Sat 21 Nov, 13:40–14:00 · Room B

U.S. income underreporting is large, persistent, and concentrated in self-employment business income, which lacks third-party verification. This paper quantifies how the In- ternal Revenue Service can close the self-employment tax gap under a fixed enforcement budget. We develop a heterogeneous-agent overlapping-generations model with occu- pational choice, endogenous business income underreporting, and risk-based auditing. Our results show that redirecting audits toward higher-risk taxpayers raises revenue by approximately 4.1%, whereas targeting high-reported-income taxpayers backfires. By endogenising evasion and enforcement, we show that statutory tax progressivity and enforcement progressivity are complements in shaping fiscal capacity.

Redistribution through Consumption Taxes: Evaluating Japan’s Dual-rate Policy
Fei Gao (Waseda University)
Session 4B · Sat 21 Nov, 14:00–14:20 · Room B

Food is the most common target of reduced consumption tax rates, but its distributional value depends on household spending patterns and the surrounding fiscal system. This paper asks how far Japan should lower its food consumption tax rate, taking the 2019 income-tax and transfer architecture as given and requiring revenue neutrality within the consumption tax. Four decades of Japanese data show a stable yen-per-calorie gradient: lower-income households devote a larger share of income to food and pay less per calorie, so food-rate relief is larger relative to income at the bottom of the distribution. I derive a closed-form condition linking the progressivity of a dual-rate consumption tax to the food Engel curve and the average propensity to consume, and then quantify welfare in a heterogeneous-agent general-equilibrium model. In the baseline third-best exercise, the welfare-maximizing policy sets the food rate at -5.5% and the non-food rate at 14.1%, yielding a consumption-equivalent gain of 0.132% relative to the 2019 system. The incidence result is stable, but the exact welfare prescription is conditional on labor-supply responses and on which other fiscal instruments may adjust. The numerical recommendation therefore concerns reform within Japan’s existing fiscal system, not the jointly optimal design of all taxes and transfers.

Digitalization, Entrepreneurship, and Wealth Inequality
Ryohei Koganemaru (The University of Tokyo); with Ichiro Muto, Fumitaka Nakamura, Makoto Nirei
Session 4B · Sat 21 Nov, 14:20–14:40 · Room B

The recent rise in wealth concentration at the top of the U.S. distribution has coincided with the ascent of digital entrepreneurs whose fortunes are tied to risky, scalable businesses built on digital technologies. We develop a tractable heterogeneous-agent general equilibrium model in which heterogeneity in risk aversion governs risky digital investment and aggregate digital capital substitutes for labor and complements traditional capital. Risk-tolerant households become digital entrepreneurs and gain disproportionately from higher digital returns, whereas risk-averse households largely rely on labor income and safe asset. As entrepreneurial wealth grows, aggregate risk tolerance—a sufficient statistic for how wealth composition affects digital capital supply—rises and further reinforces digital investment. Calibrated to U.S. data for 1989–2019, the model explains roughly half the rise in the top 1 percent share and more than half for the top 0.1 percent. Digitalization raises output and real wages but lowers the labor share, producing positive yet sharply unequal gains across households. During the transition, prices and portfolios adjust immediately, aggregate wealth gradually, and wealth composition shifts toward digital entrepreneurs over decades. Different adjustment paths of capital returns make the risk premium hump-shaped, while the slow wealth shift sustains digital deepening after excess returns decline. Entrepreneurial wealth is thus both an outcome and a propagation force of digitalization.

Estimating the Relative Contributions of Supply and Demand Drivers to Inflation in Japan
Wataru Hagio (Hitotsubashi University and Bank of Japan)
Session 4C · Sat 21 Nov, 13:40–14:00 · Room C

As in many other countries, inflation in Japan has risen rapidly and persistently since 2022. What roles have supply and demand drivers played in shaping recent inflation dynamics? To address this question, I use monthly, disaggregated, category-level data on quantities, prices, and expenditures to estimate the relative contributions of supply and demand drivers, following Shapiro (2026). This approach reveals an intriguing role for supply and demand in shaping inflation dynamics in Japan, consistent with historical developments since the late 1990s. Local projection analysis reveals that monetary tightening shocks reduce demand-driven inflation, whereas negative oil supply news shocks and exchange rate depreciation shocks increase supply-driven inflation.

Business Cycle Asymmetry According to MANK: A Mental Accounts New Keynesian ModeL
Shahmeer Hossain (University of Sydney); with James Graham, Andrej Mijakovic
Session 4C · Sat 21 Nov, 14:00–14:20 · Room C

Business cycles are known to exhibit substantial asymmetry. Recessions are typically characterized by short, sharp declines in economic activity, whereas expansions are associated with longer and more moderate increases. We investigate whether asymmetry in households’ marginal propensities to consume (MPCs) out of negative and positive income shocks can contribute to these aggregate dynamics. To model this behavior, we introduce mental accounting preferences under which households treat income and assets as non-fungible and are reluctant to draw down accumulated savings. These preferences generate substantially larger household MPCs following negative shocks. When embedded in a two-asset heterogeneous-agent New Keynesian model, they also produce larger aggregate consumption and output responses to negative rather than positive productivity shocks.

Aging, Structural Change and Service Transformation
Yasutaka Ogawa (Bank of Japan; PhD student, Keio University); with Ippei Fujiwara
Session 4C · Sat 21 Nov, 14:20–14:40 · Room C

Population aging is widely expected to shift consumption toward services. We show that in Japan, the economy with the longest and most pronounced experience of aging, it does not. We estimate an extension of the Almost Ideal Demand System of Deaton and Muellbauer (1980) using Japanese household panel data, and decompose the movement in expenditure shares since 2009 into relative-price (Baumol), real-expenditure (Engel), aging, and pandemic components. The aging effect operates through the life cycle of expenditure. Education and transportation absorb a large share of the budget while the household head is of working age and raising children, and that share recedes once those obligations end; healthcare absorbs a share that rises with age throughout. A population in which successive cohorts pass into old age therefore inherits both movements at once. Aging accordingly raises the budget shares of food (including meals away from home) and of healthcare, and lowers those of education and of transportation and communication. Because these sub-sectoral movements are of opposite sign, they cancel one another under aggregation, so that aging appears almost irrelevant at the level of the three conventional sectors. Added up in absolute value rather than allowed to net out, however, they account for roughly 30% of the total reallocation of the Japanese consumption basket across sub-sectors since 2009. Aging therefore does not shift expenditure monotonically toward services. It reallocates expenditure within them, first away from education and transportation and only later toward healthcare, and the conventional three-sector partition conceals both movements.

The Time Profile of Innovation: TFP Dynamics and Risk Premia
Xian Chen (University of Western Australia); with Shawn Xiaoguang Chen
Session 4D · Sat 21 Nov, 13:40–14:00 · Room D

This paper provides novel evidence on how patented innovation shapes the nonmonotonic time profile of firms’ productivity and the associated dynamics of equity risk premia. Using panel data on 1,203 Chinese publicly listed firms from 2005-2019 and quasi-experimental variation from China’s Strategic Emerging Industries (SEI) initiative in 2011, we document a systematic temporal asymmetry between real and financial responses to innovation. While total factor productivity (TFP) follows a hump-shaped trajectory, peaking approximately three years after a patent grant, firms’ equity risk premia increase only after productivity growth begins to decelerate, reaching a peak roughly six years after the initial shock. This pattern is consistent with a simple Bayesian learning framework in which investors infer underlying productivity dynamics from noisy outcomes when the deterministic component of TFP evolves non-monotonically.

Digital Financial Inclusion and Household Consumption: Evidence from Chinese City-Level Panel Data
Chenmiao Li (The University of Tokyo); with Xiangnan Feng
Session 4D · Sat 21 Nov, 14:00–14:20 · Room D

Using a panel of 298 Chinese prefecture-level cities from 2011 to 2023, we find that digital financial inclusion is positively associated with retail sales after city and year fixed effects. The preferred estimate is 0.0061 log points per unit of the Peking University Digital Financial Inclusion Index. A Bayesian structural equation model treats coverage, usage depth, and digitization as indicators of latent usable financial access and also yields a positive conditional relation with retail sales. We align these macro results with a component-level reconstruction of KVW-style hand-to-mouth (HtM) status from six China Household Finance Survey waves. Benchmark HtM shares range from 24.7 to 34.4 percent, low-income borrowers rely disproportionately on informal channels, and the 2019–2021 balanced panel displays substantial entry into and exit from HtM status. However, household fixed-effects estimates do not show a precise relation between DFIIC—including its usage-depth sub-index—and the observed-limit KVW-style HtM indicator. We therefore develop a dynamic-state TANK-style consumption block in which digital finance lowers payment and financing frictions within the constrained state, while the HtM share is a dynamic exposure weight rather than an imposed decreasing function of digital finance. The paper’s contribution is an internally consistent macro–micro interpretation: city-level spending covaries positively with digital financial access, but the data do not support attributing that relation to a calibrated decline in the population HtM share.

Growth and (Re)distribution in China and Korea: Bayesian Estimation of a Non-Stationary Heterogeneous Agent Model
Wending Liu (Australian National University)
Session 4D · Sat 21 Nov, 14:20–14:40 · Room D

I analyze growth and (re)distribution jointly in China (1982–2019) and Korea (1962– 2019) inside a non-stationary heterogeneous-agent model. Two innovations make Bayesian estimation feasible: a certainty-equivalent-anchored quasi-rational equilibrium concept which replaces full rational expectations with a finite-horizon deterministic forecast; and a deep-learning pipeline that amortizes the agent’s decision problem, the equilibrium computation, and the posterior itself. Counterfactual experiments find that TFP catch-up is the dominant driver of output growth yet leaves the wealth distribution untouched, that the education expansion and the fertility decline likewise buy per-capita growth with little distributional footprint, that state investment promotes growth while potentially concentrating wealth, that a thin social safety net buys growth at a large cost to consumption at the bottom, and that among the model’s estimated primitives idiosyncratic labor-productivity risk has the largest marginal impact on wealth inequality.

Asymmetric Credit Access, Inflation, and Unemployment
Hyungsuk Lee (Bank of Korea); with Sam Ng
Session 5A · Sat 21 Nov, 15:10–15:40 · Room A

We study how an empirically relevant asymmetry in credit access between employed and unemployed workers shapes the long-run relationship between inflation and unemployment. We add this asymmetry to a labour-market matching model with money in the framework of Berentsen, Menzio, and Wright (2011): employed workers settle a fraction of their goods-market trades on credit while unemployed workers transact in cash. Inflation acts through two channels of opposite sign: a Liquidity Channel that erodes firm profitability as cash trades shrink (raising unemployment), and an Employment Channel that raises the value of being employed relative to being unemployed as the cash-use asymmetry widens (lowering unemployment). At the calibrated parameters the two channels combine into a hump-shaped inflation unemployment relationship: unemployment rises with inflation at low rates and falls at high rates. Three extensions—endogenous credit adoption, common credit access, and heterogeneous per-trade credit frictions—corroborate this mechanism.

International Trade and Liquidity under Dollar Dominance
Ayushi Bajaj (Monash University)
Session 5A · Sat 21 Nov, 15:40–16:10 · Room A

The US dollar and short-term US Treasury debt remain the dominant media of exchange for international transactions. This paper introduces a dominant medium of exchange into an otherwise standard international trade model a la Melitz (2003). We show the asymmetric macroeconomic consequences of increasing trade barriers. While higher trade costs lead to a lower share of exporters and productivity losses across the board, we find that the dominant country experiences a steeper decline in consumption and welfare compared to its trading partner. The fall in demand for an internationally accepted medium of exchange causes its price to fall which adversely affects the dominant country’s overall welfare.

Liquidity, Market Power and Capital Accumulation
Sam Ng (University of Macau)
Session 5A · Sat 21 Nov, 16:10–16:40 · Room A

We develop a monetary model in which greater market power does not necessarily worsen economic welfare. We endogenize two dimensions of market structure: bank entry, which determines access to banking services, and sellers’ costly investment in rent extraction, which determines their share of trading surplus. Greater seller market power distorts consumption by raising buyers’ payments, but also strengthens private incentives to accumulate capital, mitigating underaccumulation in the decentralized economy. Banking shapes this trade-off by relaxing liquidity constraints and changing the composition of money- and credit-financed transactions. We summarize these forces through a capital premium associated with production in the frictional goods market. We show that when bank entry and seller markup incentives respond to monetary policy, a positive policy rate can reduce capital inefficiency. However, this also comes at the cost of higher goods markups and lower consumption. Thus, there is a trade-off that can generate non-monotone welfare effects of monetary policy. Our main insight is that market power is not necessarily welfare detrimental. It can partially correct decentralized capital underaccumulation. Eliminating the conventional monetary friction therefore need not maximize welfare when it also weakens the private incentives to invest in productive capital.

Stock Market Participation, Liquidity, and Monetary Policy
Lu Wang (University of Western Australia); with Aaron Sullivan, Sarah Xiao
Session 5A · Sat 21 Nov, 16:40–17:10 · Room A

We study how monetary policy shapes stock market participation, and how the participation margin in turn shapes asset prices and the transmission of policy. We develop a New Monetarist model in which households may purchase an asset but must pay a fixed cost in order to do so. We show that the effect of the nominal interest rate on stock market participation is state-dependent. That is, monetary policy becomes less effective as it gets tighter, consistent with empirical evidence. When stock market tightness is endogenous, multiplicity emerges for sufficiently high nominal interest rates, and households’ asset market participation exhibits inertia and hysteresis.

Gender Gaps in Marriage Effects and Human Capital Accumulation
Liyuan Dong (New York University Abu Dhabi)
Session 5B · Sat 21 Nov, 15:10–15:40 · Room B

This paper examines how marriage affects the labor incomes of men versus women over the life cycle. Using data from the 1985 to 2018 Panel Study of Income Dynamics (PSID), I document that marriage increases wage rates and working hours for men but reduces both for women, with the effects most pronounced among highly educated men and less educated women. Based on these facts, I propose the mechanism of household specialization and human capital accumulation to explain the contrasting results. I develop a dynamic life-cycle model in which two household members jointly make labor supply decisions and human capital is accumulated through learning-by-doing. Household specialization and marketization of home production together explain the heterogeneous effects of marriage across gender and education groups.

Career Derailment: Occupation-Switching Costs and Gendered Misallocation of Talent
Michelle Petersen Rendall (Monash University); with Kanika Goyal, Asha Sundaram
Session 5B · Sat 21 Nov, 15:40–16:10 · Room B

Substantial gender inequality persists in earnings and work despite convergence in education and labor market entry. This paper studies how the dynamics of reallocation contribute this persistence. When structural change displaces workers, women face different occupational switching costs partially driven by the gendered opportunity cost of home production. Displacement lowers the opportunity cost of domestic work, inducing some women (men) to withdraw from market work rather than bear the cost of occupational transition. Combined with the large and persistent child penalty upon labor-market re-entry, this response generates gendered misallocation of talent with aggregate productivity consequences. We develop a quantitative general equilibrium model that features multi-sector occupational choice, costly and irreversible occupation switching, and endogenous family-related labor-force withdrawal. We use the China trade shock as an instrument for labor-demand displacement, exploiting cross-sectoral variation in exposure, to identify the causal effect of displacement on gendered occupational transitions and household decisions in estimating and disciplining the model.

Gender Gaps in Time Use and Entrepreneurship
Faisal Sohail (University of Melbourne); with Pedro Bento, Lin Shao
Session 5B · Sat 21 Nov, 16:10–16:40 · Room B

The prevalence of entrepreneurs, particularly low-productivity self-employed, declines as economies develop. This decline is more pronounced for women: relative to men, they are more likely to be entrepreneurs in poor economies but less likely in rich ones. We investigate whether gender gaps in time devoted to non-market activities, which narrow with development, can account for this pattern. We develop a quantitative framework in which selection into occupations depends on ability and time, with features gender-specific distortions and social norms that influence market participation and the allocation of non-market time within the household. Calibrated to match cross-country data, the model finds differences in social norms to be primarily responsible for gender gaps in time use, and, through them, for two-thirds of the level of gender gaps in self-employment across countries. Through these channels, social norms shape the quantity and quality of businesses in an economy, with meaningful implications for aggregate output.

Tax Cuts by Occupation: Entrepreneurs vs. Workers
Eunseong Ma (Yonsei University); with Myeongju Kim
Session 5B · Sat 21 Nov, 16:40–17:10 · Room B

This paper studies how the macroeconomic effects of tax cuts depend on occupational targeting—toward entrepreneurs versus wage workers. Using a new state-level panel of occupation- specific federal tax shocks for the United States from 1981 to 2017, we find that entrepreneur- targeted tax cuts generate substantially larger increases in output, consumption, and employment than revenue-equivalent worker-targeted cuts. These effects coincide with increases in both entrepreneurship and wage employment, pointing to business formation and firm expansion as key transmission channels. An incomplete-markets model with occupational choice attributes these findings to earnings-based borrowing constraints and demand-driven amplification that jointly produce large entrepreneur multipliers.

Financial Development, Competition, and Productivity
Marcel Peruffo (University of Sydney)
Session 5C · Sat 21 Nov, 15:10–15:40 · Room C

I study how financial development promotes competition. I extend the model of Atkeson and Burstein (2008) to include dynamic capital accumulation and financial frictions. Capital market imperfections interact with endogenous misallocation from market power and variable markups. Calibrating the model with data from Chilean manufacturing firms, I show that improved access to external finance reduces industry concentration and both the level and dispersion of markups. These effects explain over half of the total productivity gains from financial development and are especially pronounced in financially underdeveloped economies. Models that ignore markup variation thus underestimate the benefits of improving capital markets.

On Markups, Borrowing Constraints and Firm Size
Begoña Domínguez (University of Queensland); with Andres Bellofatto, Patrick Elkington, Alicia Rambaldi
Session 5C · Sat 21 Nov, 15:40–16:10 · Room C

This paper studies the relationship between firm-level borrowing constraints and markups. Using comprehensive administrative data on businesses operating in Australia, we proxy borrowing-constraint tightness by combining managers’ survey responses on financial decision-making, and firms’ balance sheet data from business income tax returns. Within narrowly defined industries, we find that firms facing tighter borrowing constraints tend to charge considerably higher markups. This relationship is negatively related to firm size: at the median of the size distribution, constrained firms set markups about 3 percent higher than unconstrained firms, but such difference can exceed 10 percent among the smallest firms. To interpret these findings, we develop a model in which working capital requirements generate a positive link between borrowing constraints and markups. Consistent with our theory, we further show empirically that higher working capital requirements substantially amplify the pass-through from borrowing constraints to markups.

Smart Contracts, Prepayment and Seller Selection: large Evidence from China’s Education-Training Industry
Sylvia Xiaolin Xiao (Peking University); with Boyan Jovanovic
Session 5C · Sat 21 Nov, 16:10–16:40 · Room C

We study how smart contracts embedded in digital payments reshape markets that run on prepayment, in an equilibrium model where sellers differ privately in reliability and buyers prepay under one of three contracts: money up front (M), smart-contract escrow (S), or platform escrow (I). It is well known that China has the world’s largest sovereign CBDC pilot and the only retail CBDC still active at scale. We map the theory to China’s education-training industry and e-CNY pilot in one of the largest cities in China, which by July 2026 had locked roughly 0.9 million prepaid-tuition contracts and RMB 4.5 billion of tuition under smart contract escrow. Three main results follow. First, contract choice screens sellers: reliable types adopt escrow, unreliable types take the money up front, and delivery-contingent competition can unravel the pooling that sustains up-front payment. Second, learning by doing lowers adoption costs, generating a diffusion path along which smart contracts spread and aggregate default falls, while the unescrowed segment worsens in composition. Third, the public escrow rail caps the platform’s sustainable commission at the value of its matching service. To our knowledge, this is the first industrial-organization study of CBDC-embedded smart-contract adoption that combines an equilibrium model with empirical evidences.

The Limit of Credit Policy and Its Fiscal Remedy
Yunpeng Deng (UNSW Sydney)
Session 5C · Sat 21 Nov, 16:40–17:10 · Room C

Theory often suggests that credit policy should stabilize credit spreads during a financial downturn. We show that this recommendation may have a limit. In a model where banks’ leverage is forward-looking and self-referential, general- equilibrium feedback through credit spreads is essential to anchor banks’ ex- pectations about their own future leverage. Stabilizing the credit spread too aggressively short-circuits this feedback, leaving banks’ asset demand and the central bank’s own balance sheet indeterminate. We characterize this limit an- alytically and show that it exists in several published studies. An active fiscal policy offers a remedy: it anchors banks’ expectations by restricting the feasible path of the central bank’s balance sheet.

Optimal Innovation under Population Decline: Growth and Firm Dynamics
Minsu Chang (Seoul National University); with Toshihiko Mukoyama
Session 5D · Sat 21 Nov, 15:10–15:40 · Room D

How should innovation policy respond to population decline? We first document that establishment dynamics relate to working-age population growth across U.S. states and across Japanese prefectures. We then build an endogenous growth model with three innovation margins: new product entry, creative destruction, and own-good improvement. In the model equilibrium, population growth affects the aggregate outcome through the change in the value of innovation. Slower population growth reduces entry and exit, and tilts innovation toward incumbents. The first-best allocation is decentralized by taxing creative destruction, taxing entry, and subsidizing production at the same time; the creative-destruction tax and entry tax are independent of demographics, but the optimal production subsidy decreases as population growth falls.

Innovation, Diffusion, and Misallocation: Understanding China’s Recent Productivity Dynamics
Yan Liang (Deakin University); with Emin Gahramanov, Xueli Tang, Chengsi Zhang
Session 5D · Sat 21 Nov, 15:40–16:10 · Room D

Since the 2008 financial crisis, China has experienced a marked slowdown in productivity growth despite exceptionally strong innovation activity. We study this puzzle using an endogenous growth model calibrated to the Chinese economy over 2007–2017. The analysis identified two central forces. First, innovation has become increasingly costly as China approaches the technological frontier. Second, intensified involutionary competition diverts innovation effort towards activities with limited social gains. Other factors, including weaker knowledge diffusion and greater resource misallocation, play a minor role. While industry policies support innovation, they are insufficient to offset these broader structural forces. Sustaining productivity growth therefore requires improving efficiency and direction of innovation activity, particularly by curbing involutionary competition.

What Does an Imported Good Teach?
Zoe Zhang (The University of Tokyo)
Session 5D · Sat 21 Nov, 16:10–16:40 · Room D

What knowledge do firms acquire through imports? I compare learning from a source country’s knowledge pool, its product-specific expertise, and the firms that export to the buyer’s market. In a dynamic Melitz model, the first two mechanisms attach the same knowledge distribution to a source across destinations. Exporter selection instead makes that distribution depend on the bilateral export cutoff: admitting less productive exporters reduces knowledge per learning opportunity. Chinese customs records linked to importing firms’ patent citations support a product-specific component. Conditional on total imports from a country, firms cite that country more in the technologies represented in their own import basket. This relationship holds in two samples covering patent years 2001–2007 and 2008–2014. The negative relationship with import penetration predicted by the conditional Pareto selection benchmark receives little support. The findings suggest that the product composition of trade matters for knowledge diffusion, beyond the countries from which firms buy.

Optimal Taxation in the Age of AI Uncertainty
Pei-Cheng Yu (UNSW Sydney); with Radoslaw Paluszynski, Anastasios Karantounias
Session 5D · Sat 21 Nov, 16:40–17:10 · Room D

This paper examines the impact of non-insurable technological shocks on labor markets and optimal tax policy. Using survey data, we document a substantial dispersion in artificial intelligence (AI) pessimism that is negatively correlated with worker income. Motivated by these findings, we build a dynamic Mirrlees income taxation model with ambiguity-averse workers. Workers form endogenous worst-case beliefs about the impact of new technologies, that vary by skill level. This endogenous heterogeneity in expectations enables the formulation of government policies that indirectly influence workers’ subjective beliefs to enhance welfare. We calibrate the model to the empirical findings on AI expectations and solve for optimal tax schedules and AI investment. We find that uncertainty over AI motivates a more progressive labor income tax schedule, combined with a mostly flat savings subsidy of about 12%. The optimal policy reverses the sign of the empirical relationship between AI pessimism and worker income.

Parenting Style and Fertility: A Cross-country Analysis
Siew Ling Yew (Monash University); with Shuyun May Li, Solmaz Moslehi, Yao Yao
Session 6B · Sun 22 Nov, 09:30–10:00 · Room B

Parenting style and fertility are inherently interrelated. Intensive parenting increases parental effort and discourages fertility, while having more children constrains the time and effort parents can devote to each child, shaping parenting choices. Despite this interaction, existing research has largely studied fertility and parenting style in isolation. This paper bridges the two literatures by examining how socioeconomic conditions and policies jointly shape parents’ choices of fertility and parenting style, and how parenting style mediates fertility outcomes. Using data from the World Values Survey, we document systematic cross-country patterns in parenting styles and fertility across stages of economic development. As income rises, parenting shifts from authoritarian to authoritative and eventually toward permissive styles, while fertility declines and partially rebounds in high-income economies. Micro-level evidence confirms that, relative to authoritative parents, authoritarian parents have more children, whereas permissive parents have fewer. To interpret these patterns, we develop a quantitative overlapping-generations model in which parents jointly choose fertility and parenting style in response to socioeconomic environments, including returns to education, inequality, structural change, and the marketisation of childcare. The model is calibrated to a benchmark economy, US, and to a selective group of countries at different development stages to account for observed patterns in fertility and parenting style within and across countries. Our quantitative analysis aims to uncover the role of parenting style in the quantity–quality tradeoff underlying fertility decisions, evaluates the contribution of socioeconomic and macro factors, and assesses policies aimed at raising fertility.

College as Human Capital Investment or Tournament: A Macroeconomic Analysis
Siyu Shi (New York University Abu Dhabi)
Session 6B · Sun 22 Nov, 10:00–10:30 · Room B

A college education is commonly viewed as a productive investment, but it also serves a competitive purpose by allowing individuals to stand out in the job market, as in a tournament. I study the relative importance of these two channels in determining college attendance. I build a general equilibrium life-cycle model with a college decision, human capital investments, and skill allocation. The model’s novelty is that workers are allocated to different occupations based on their relative ranking of accumulated human capital, so a college education also has competitive value. Parameterizing the model to the U.S. economy, I find that the competitive channel accounts for 27% of college attendance but encourages overinvestment in human capital accumulation. The optimal policy lowers college subsidies and tax progressivity, improving welfare by 0.9% by mitigating competition in college education and its negative externalities. The results are also robust when subsidies are reformed to be wealth- or merit-based.

Saving by Shopping? The Limits of Substitution in Household Production
Kozo Ueda (Waseda University); with Miyu Iwasaki
Session 6B · Sun 22 Nov, 10:30–11:00 · Room B

To what extent can households buffer cost-of-living shocks by shopping more efficiently? Using novel retailer scanner data linked to individual shoppers at a single supermarket, we isolate the savings generated by the intensive margin of shopping. We develop a structural model of endogenous shopping frequency and estimate the elasticity of substitution between time and market goods implied by this intensive margin to be approximately 0.3, which is well below the values exceeding unity typically assumed in macroeconomics. This suggests that the within-store timing margin alone affords households only limited scope to smooth consumption through shopping effort. Welfare analysis shows that increases in fixed time costs lead to sizable welfare losses.

Tenancy Without Rent, Credit Without Banks: Housing as Repo in Korea
Yunho Cho (Sogang University); with Yoonsoo Lee, Hyeonuk Shin
Session 6C · Sun 22 Nov, 09:30–10:00 · Room C

This paper studies a housing finance system in which tenants provide credit directly to landlords. In Korea, this arrangement is known as jeonse. Tenants transfer a large refundable deposit in place of rent, allowing landlords to finance housing investment with tenant funds. To quantify the implications of jeonse, we develop a general equilibrium heterogeneous-agent life-cycle model in which households choose among homeownership, jeonse, and periodic rent. The model features subjective beliefs about future house prices, which governs expected capital gains and landlords’ willingness to supply jeonse housing. We calibrate the model to Korean micro-data and discipline this belief using observed rates of jeonse and rental supply. The model shows that deposit financing alone cannot sustain jeonse without sufficiently optimistic price beliefs. Eliminating jeonse reduces deposit-financed housing investment but redirects tenants toward homeownership and periodic rent. The reallocation effect dominates, raising house prices by 15.8% and rents by 18.8%. Relaxing jeonse loan limits expands young households’ access to jeonse and later homeownership, but also increases landlord financing and raises prices and rents. Jeonse therefore expands housing access while providing belief-driven leverage to housing investors.

Competition, risk, and pricing in the US Mortgage Market
Mei Dong (University of Melbourne); with Eungwon Lee, May Li
Session 6C · Sun 22 Nov, 10:00–10:30 · Room C

We study how banking competition affects mortgage pricing in the mortgage market. Using the US mortgage data from 2014 to 2019, we find that greater competition in the local mortgage market is associated with higher mortgage rates and riskier borrower pools. These relationships are nonlinear: both effects are stronger in markets that are more competitive. Moreover, regions in which borrower risk responds more to competition exhibit larger increases in mortgage rates. We develop a model where lenders choose both loan supply and loan risk while engaging in Cournot competition in loan and funding markets. Competition lowers the loan rate through the conventional Cournot channel but also induces greater risk-taking. When competition is sufficiently intense, the risk effect dominates the Cournot effect, so further competition raises rather than lowers mortgage rates. These results highlight a risk-taking channel through which greater banking competition can increase borrowing costs.

The Political Economy of Government Mortgage Guarantees
Jiheum Yeon (Korea Institute for International Economic Policy); with Dong Ho Kang
Session 6C · Sun 22 Nov, 10:30–11:00 · Room C

Government insurance of mortgage credit risk has survived decades of reform, even though normative analyses find that it lowers aggregate welfare. To understand why the policy that is realized diverges from the policy that is optimal, we develop a quantitative housing model in which the guarantee fee is selected by a political criterion. The economy is populated by groups with opposing stakes in mortgage policy and different degrees of patience: impatient borrowers, who finance housing with defaultable long-term mortgages and choose between owning and renting, and patient savers and landlords. Raising the fee is passed through to the mortgage rate, lowers house prices and the housing stock, and redistributes permanently from borrowers to savers and landlords. Whether the fee is raised therefore depends on the horizon over which each group’s stake is capitalized. A planner who values claims over an infinite horizon adopts the highest admissible fee, whereas a policymaker accountable over any empirically relevant horizon retains the low fee: the horizon at which the two criteria agree is roughly 27 years, an order of magnitude longer than an electoral term. The subsidy persists not because it is efficient, but because those who would lose from its removal are impatient and hold a large stake.

Fixed Costs, Product Heterogeneity, and The Force of Competition
Maria Ptashkina (University of Melbourne); with Vladimir Asriyan, Alberto Martin, Jaume Ventura
Session 6D · Sun 22 Nov, 09:30–10:00 · Room D

One of the most remarkable facts of the US economy over the last few decades is the rise of fixed costs. Indeed, the fixed-cost share (ratio of fixed to total costs) of the US economy has increased significantly during this period, as has its dispersion across firms. This paper provides a framework to think about the macroeconomic origin and effects of fixed costs. The core of the framework is that goods are heterogeneous, as they differ in their demand elasticities, as well as in their fixed and variable costs. We show that, in a contestable equilibrium, the economy’s fixed-costs share is determined endogenously both by the extensive margin, i.e., the set of goods being produced, and the intensive margin, i.e., the quantity produced of each good. These two margins also shape the effects of economic growth in the fixed-cost share: while intensive growth tends to reduce the fixedcost share by raising the number of units produced, extensive growth tends to increase the fixed-cost share by activating goods with increasingly high fixed-costs. We use a numerical exploration of the model to rationalize the evolution of fixed costs in the US, and perform a normative analysis of the fixed-cost economy to characterize the distortions that affect the intensive and extensive margins of production in the contestable equilibrium.

Propagation over Production Networks with Endogenous Market Power
Daniel Prosi (New York University Abu Dhabi)
Session 6D · Sun 22 Nov, 10:00–10:30 · Room D

This paper studies how changes to firm productivity or size propagate over a production network. It derives an endogenous measure of market power when firms internalise their position in the production network to extract monopoly rents. The paper first explores the relation between network position and endogenous markups under general non-parametric production functions. This separates economically meaningful channels of interaction from restrictions introduced by the choice of functional form of production and the competition regime. It then studies how changes to productivity or firm-size propagate over the network when markups are endogenous. The forward equation of productivity gains and the backward equation of demand creation are complemented with a horizontal equation that captures changes in endogenous market power. Using granular administrative data from Rwanda and an instrumental variable strategy based on exogenous border closures, the paper tests the empirical validity of the derived measures. It then uses the universe of firm-to-firm interactions in the country to study the network-position of Foreign Direct Investment (FDI) in Rwanda and assess the general equilibrium effects of FDI on the domestic economy via efficiency gains, demand creation, and changes to the structure of competition.

Self-Fulfilling Sentiment(s) in Production Networks
Weimin Zhou (Sichuan University); with Renbin Zhang, Jinhui Xu, Zijian Wang
Session 6D · Sun 22 Nov, 10:30–11:00 · Room D

Building on the sentiment mechanism of Benhabib et al. (2015), we study self-fulfilling price movements in a production network. Firms set prices using noisy signals that confound local costs with endogenous prices, while supplier prices enter buyers’ marginal costs. Production links therefore create an additional feedback channel through which belief-induced price responses can be inferred and retransmitted. For aggregate sentiment, production links enable a self-fulfilling price movement when firms’ common-price exposure is sufficiently aligned with their input-sourcing patterns. We then consider coexisting sectoral sentiments. With local price signals, sentiment must begin in a sector with direct support; supplier-price feedback can then transmit it downstream and allow additional sectors to sustain distinct sentiment movements. With cross-sector price signals, price comovement can provide support even when no sector sustains sentiment alone. Production links and firms’ information thus jointly determine the existence, spread, and dimensionality of sentiment.

New Keynesian Economics through the Extensive Margin
Akira Ishide (The University of Tokyo); with Saki Bigio
Session 7A · Sun 22 Nov, 11:15–11:45 · Room A

This paper reformulates the New Keynesian model to incorporate output adjustments through the extensive margin. Shifting from adjustments through the intensive to the extensive employment margin, the model introduces predetermined output, altering key properties of the New Keynesian framework. First, the Taylor principle is inverted: stability is achieved when nominal rates respond less than one-for-one with inflation. Second, the model significantly alters the output responses to changes in monetary policy. We argue that this represents a challenge and an opportunity for the literature. Sticky information allows the model to correct the sign of impulse responses.

Zero liquidity and concentrated redistribution in a New Keynesian model
Yuichiro Waki (Aoyama Gakuin University)
Session 7A · Sun 22 Nov, 11:45–12:15 · Room A

The zero-liquidity assumption — households cannot borrow and assets are in zero net supply — simplifies heterogeneous-agent models by eliminating the asset distribution as an aggregate state. I show that it creates a peculiar response to concentrated redistribution in a New Keynesian model. A transfer $\Delta$ distributed among a fraction $\varepsilon$ of households reduces period-0 output by $\Delta/\varepsilon$. Thus, output can remain depressed as $\Delta$ vanishes and can approach zero. If a positive fraction of transfer-paying households can borrow and their constraints are locally slack, the output loss vanishes with $\Delta$. This fragility warrants caution in highly concentrated transfer experiments.

Learning about monetary policy in Tractable HANK
Christopher Gibbs (University of Sydney); with Paul Kitney, Nigel McClung
Session 7A · Sun 22 Nov, 12:15–12:45 · Room A

A growing literature distils the aggregate properties of Heterogeneous Agent New Keynesian (HANK) models into parsimonious representations with the same reduced form as a Representative Agent New Keynesian (RANK) model. The monetary policy settings that induce determinate and learnable equilibrium in the latter are well-known. We revisit these findings for tractable versions of HANK models. We find that when HANK economies imply countercyclical income risk, their tractable counterparts require different monetary policy settings in order to ensure determinacy and learnability (E-stability) compared to the RANK benchmark. This includes equilibria that are indeterminate and E-stable, while the conventional Taylor principle is satisfied. Monetary policy must be history dependent or place more weight on output stabilization to match the RANK model’s properties.

Public and Self Insurance in Job Search and Consumption: Evidence from Bank Transaction Data
W. Similan Rujiwattanapong (Waseda University); with Fei Gao, Kozo Ueda
Session 7B · Sun 22 Nov, 11:15–11:45 · Room B

How do private savings and public unemployment insurance (UI) interact in shaping job transitions? This paper examines the effects of UI and private wealth on reemployment wages and consumption using novel bank transaction data covering 65,697 individuals in Japan. We develop a job-search model in which unemployed workers choose target wages and consumption while facing UI benefits and financial buffers. The model predicts that UI raises target wages, whereas the effect of wealth is theoretically ambiguous. Empirically, both UI generosity and wealth are associated with higher reemployment wages, and the effect of UI remains strong even among wealthy individuals. Consumption rises with reemployment wages and falls following job separation, particularly for individuals with limited liquidity. UI benefits substantially mitigate these consumption losses and help sustain consumption during nonemployment. Moreover, by increasing reemployment wages, UI continues to affect consumption after reemployment. The results suggest that private wealth and UI play distinct roles in shaping labor market behavior and consumption dynamics and highlight the importance of liquidity constraints in evaluating unemployment insurance policies.

Disability Benefit Reform, Employment and Welfare: Evidence from Australia’s Welfare-to-work Reforms
Man Hin Chio (Australian National University); with Chung Tran
Session 7B · Sun 22 Nov, 11:45–12:15 · Room B

We study the labour market effects of tightening eligibility for disability support among people with partial work capacity. Australia’s 2005 Welfare-to-Work reform lowered the Disability Support Pension (DSP) work capacity threshold from 30 to 15 hours per week, and grandfathering and delayed reassessment generated staggered exposure to the new rule. Using linked administrative records from the Person-Level Integrated Data Asset and a staggered difference-in-differences design, we find that labour earnings decline upon first exposure and that employment falls in subsequent quarters, with both effects persisting over time. We interpret this evidence using a lifecycle model of the Australian economy that features incomplete markets, saving, discrete labour supply, separate productivity and disability shocks, progressive income taxation, the means-tested Age Pension, and a means-tested DSP with an hours rule, income and assets tests, permanent assessed medical eligibility, and an endogenous claiming decision. Lowering the hours threshold from 30 to 15 in the calibrated model reduces average labour income, hours, and employment in the working-age population, and entrant welfare falls in every education and disability group. A simulation of 100,000 lifecycle paths reproduces the qualitative pattern of the empirical estimates for recipients who remain on and who exit the payment. Our findings suggest that policies targeting eligibility tightening can weaken insurance and generate labour market disincentive effects, reducing employment and earnings instead of promoting work among people with partial work capacity.

Redistribution, distortions, and the welfare effects of Social Security
Ponpoje Porapakkarm (National Graduate Institute for Policy Studies); with Svetlana Pashchenko, Youngsoo Jang
Session 7B · Sun 22 Nov, 12:15–12:45 · Room B

What is the best way to reform Social Security? Academic literature offers diverging advice. There is a well-known result that the optimal size of Social Security is zero, implying it is best to phase the program out. Other studies argue that much can be gained by redesigning the program, given its current size. We provide a unified analysis that examines how the optimal size of Social Security depends on the key features of its design. We first develop a theoretical decomposition tracing the program’s welfare effects to (i) income redistribution, (ii) distortions on the annuitization level, and (iii) intertemporal distortions. We then quantitatively assess the role of these channels. We show that the zero-optimal-size result arises because Social Security is too distortive and not redistributive enough. Once these design flaws are corrected, it is even optimal to increase the size of the program.

Intergenerational Spillovers: The Impact of Labor Market Risk on the Housing Market
Leanne Nam (Aarhus University)
Session 7C · Sun 22 Nov, 11:15–11:45 · Room C

Unemployment leads to large and persistent income losses for workers. Higher unemployment in the labor market therefore has spillover effects on the housing market. This paper studies such spillover effects from both empirical and theoretical perspectives. Using data from the Current Population Survey (CPS), I show that a 1 percentage point increase in the unemployment rate leads to a 1.55% decline in housing prices. Theoretically, I develop an overlapping generations model with a housing market. The calibrated model replicates the empirically observed spillover effect for the U.S. economy. Higher income uncertainty is the main driver of the spillover effect, rather than actual income losses. The spillover effect transmits one-third of the welfare losses of workers due to higher unemployment in the labor market to older, retired households by reducing their housing wealth. Younger workers benefit in part by buying houses at depressed prices. The magnitude of the spillover effect is shaped by the demographic structure of the population and the specific age groups affected by unemployment shocks. I find that increasing the generosity of unemployment insurance stabilizes the housing market, although it only partially mitigates the spillover effect.

Blame the Boomers or Foreigners? A Structural Decomposition of Housing Costs and Intergenerational Inequality.
Takeki Sunakawa (ASEAN+3 Macroeconomic Research Office); with Tina Kao, Timothy Kam
Session 7C · Sun 22 Nov, 11:45–12:15 · Room C

We decompose and quantify the different drivers of the cost of home-ownership. We show that immigration accounts for a small share, and this depends on the distribution of immigrant types.

Beyond the Floodplain: Uncovering the Spatial Spillovers of Land Price Declines after Typhoon Hagibis
Fumitaka Nakamura (Bank of Japan); with Toshitaka Maruyama, Hiroaki Shirai, Nao Sudo
Session 7C · Sun 22 Nov, 12:15–12:45 · Room C

We examine how changes in flood-risk perception influence land prices by analyzing price developments in inundated and nearby areas following Typhoon Hagibis (2019), using parcel-level data, inundation maps, and an event-study approach. Our key findings indicate that land prices in inundated areas declined monotonically after the flood, falling 10% after five years. Furthermore, a distance-decay effect is observed, with non-inundated plots within 100 meters of the flood boundary declining by 5%, whereas effects disappear beyond 800 meters. Price declines differ inside and outside official hazard zones, with larger spillover effects in unaffected areas outside the zones. To clarify the mechanism, we extend the Bayesian belief-updating framework of Bakkensen and Barrage (2022) to account for variations in flood-risk perception arising from temporal lags, proximity to the affected area, and hazard-zone designation. Simulation results based on the model show that three factors explain price changes: delayed adjustment of risk perceptions after the flood, reduced impact of flood risk perceptions with distance from the disaster area, and pre-flood incorporation of flood risk into prices within hazard map zones.

Equilibrium Indeterminacy When Only Some Firms Are Scalable
Oscar Pavlov (University of Tasmania); with Mark Weder
Session 7D · Sun 22 Nov, 11:15–11:45 · Room D

This paper develops a dynamic general equilibrium model with heterogeneous firm scalability and studies its implications for macroeconomic fluctuations. Firms with scalable production technologies endogenously account for a disproportionate share of output, employment, and market power, causing aggregate returns to scale to vary endogenously with macroeconomic activity. We show that empirically plausible differences in firm scalability are sufficient to generate equilibrium indeterminacy. The calibrated model reproduces key features of the U.S. distributions of firm size and markups, while Bayesian estimation indicates both the importance of scale heterogeneity and that animal spirits shocks account for a non-negligible share of business cycle fluctuations, particularly in investment.

Lumpy Investment Commitments and Smooth Capital Expenditure
Ikuo Takei (Asia School of Business); with Ryota Ogaki
Session 7D · Sun 22 Nov, 11:45–12:15 · Room D

Using mandatory project-level disclosures by Japanese listed firms, we observe multi-year in- vestment commitments separately from capital expenditure. Commitments are lumpy: in most firm-years firms start no new project, and a firm’s largest year carries most of its commitments. Capital expenditure is smooth, because each commitment is paid out over a multi-year build. Commitments also respond about twice as strongly as spending to cash flow and to analysts’ sales-forecast revisions, mostly through new projects. A calibrated investment-planning model with a single pipeline state attributes both gaps to time to build, not financing constraints, so spending-based measures understate investment responses.

Common Revisions, Relative Positions, and the Micro-Macro Wedge
Greta Meggiorini (University of Auckland)
Session 7D · Sun 22 Nov, 12:15–12:45 · Room D

Forecast-error-on-revision regressions suggest that consensus forecasts underreact while individual forecasts overreact. Using aligned SPF samples, I show that this contrast is unstable across variables, horizons, and sample windows. I derive an exact decomposition of the pooled individual coefficient into slopes on common revisions and changes in forecasters’ positions relative to consensus, weighted by revision synchronization. The common slope is regime-sensitive. The relative-position slope is more stable and negative, but stable relative-position dispersion mechanically implies a benchmark of -1/2, which most output and price estimates do not reject. The individual coefficient is therefore not a standalone test of behavioral bias.