We develop a neural-network framework for multi-period risk--reward stochastic control problems with constrained two-step feedback policies that may be discontinuous in the state. We allow a broad class of objectives built on a finite-dimensional performance vector, including terminal and path-dependent statistics, with risk functionals admitting auxiliary-variable optimization representations (e.g.\ Conditional Value-at-Risk and buffered probability of exceedance) and optional moment dependence. Our approach parametrizes the two-step policy using two coupled feedforward networks with constraint-enforcing output layers, reducing the constrained control problem to unconstrained training over network parameters. Under mild regularity conditions, we prove that the empirical optimum of the NN-parametrized objective converges in probability to the true optimal value as network capacity and training sample size increase. The proof is modular, separating policy approximation, propagation through the controlled recursion, and preservation under the scalarized risk--reward objective. Numerical experiments confirm the predicted convergence-in-probability behavior, show close agreement between learned and reference control heat maps, and demonstrate out-of-sample robustness on a large independent scenario set.

This study explored the association between sleep duration and redistribution preferences. Using an online survey, we propose a hypothetical situation in which the tax paid directly by respondents is redistributed to those earning less than one-fifth of the respondents' income. Next, we asked about the allowable tax rates. We found the following through Tobit and ordered logit regression estimations: (1) The relationship between sleep hours and the allowable tax rate showed an inverted U-shape, where the optimal amount of sleep led to the highest allowable tax rate. (2) High-quality sleep was more positively correlated with the allowable tax rate than was low-quality sleep when the sleep quantity was the same. (3) Sleep hours were more significantly and positively correlated with the allowable tax rate in the high-income group than in the low-income group. (4) Assuming that twice the amount of tax paid goes to those with lower income, individuals who previously preferred a higher tax rate were more likely to increase the allowable tax rate.

Using an individual-level panel dataset from Japan covering the period 2016-2024, we examined how the COVID-19 pandemic, as an unanticipated public crisis, affected preferences for income redistribution. Furthermore, we investigated how the association between redistribution preferences and trust in government changed before and after COVID-19. The major findings are as follows: (1) individuals in the high-income group are less likely to prefer redistribution after COVID-19 than before it; (2) the degree of decline in redistribution preference is lower when trust in government is higher; and (3) generalised trust and reciprocity did not influence the decline in preference.

This study investigates shifts in acceptable tax rate for reducing inequality during the COVID-19 pandemic using Japanese data. We find a transition from norm-based, unconditional support for redistribution to conditional altruism. Before the pandemic, support remained high and independent of institutional trust. The pandemic generated an overall decline in altruistic attitudes while increasing their dependence on trust in government, particularly among high-income individuals. This "widening gap" implies that in post-crisis societies, the social contract is no longer anchored in stable social norms but increasingly relies on institutional trust to sustain income redistribution from the rich to the poor.

Leveraged ETFs (L-ETFs) are exchange-traded funds that achieve price movements several times greater than an index by holding index-linked futures such as Nikkei Stock Average Index futures. It is known that when the price of an L-ETF falls, the L-ETF uses the liquidity of futures to limit the decline through arbitrage trading. Conversely, when the price of a futures contract falls, the futures contract uses the liquidity of the L-ETF to limit its decline. However, the impact of arbitrage trading on the liquidity of these markets has been little studied. Therefore, the present study used artificial market simulations to investigate how the liquidity (Volume, SellDepth, BuyDepth, Tightness) of both markets changes when prices plummet in either (i.e., the L-ETF or futures market), depending on the presence or absence of arbitrage trading. As a result, it was found that when erroneous orders occur in the L-ETF market, the existence of arbitrage trading causes liquidity to be supplied from the futures market to the L-ETF market in terms of SellDepth and Tightness. When erroneous orders occur in the futures market, the existence of arbitrage trading causes liquidity to be supplied from the L-ETF market to the futures market in terms of SellDepth and Tightness, and liquidity to be supplied from the futures market to the L-ETF market in terms of Volume. We also analyzed the internal market mechanisms that led to these results.

We establish an existence of equilibrium result for a class of non-Markovian mean-field games with unbounded control space in weak formulation. Our result is based on new existence and stability results for quadratic-growth generalized McKean-Vlasov BSDEs. Unlike earlier approaches, our approach does not require boundedness assumptions on the model parameters or time horizons and allows for running costs that are quadratic in the control variable.

This paper develops a nonlinear theoretical framework to analyze the dynamics of public expenditure reallocation in Uruguay. Motivated by recent debates on fiscal reform and expenditure efficiency, the paper models fiscal adjustment as a dynamic process in which expenditure categories exhibit heterogeneous institutional rigidity and convex adjustment costs. Using the national budget for the 2026-2030 fiscal period as an institutional reference, the paper presents a calibrated illustration of the theoretical framework that captures key features of the structure of public spending, including transfers, the public wage bill, operating expenditures, and public investment. The calibration translates institutional characteristics of the budget into quantitative transition dynamics rather than estimating structural parameters econometrically. The framework allows the evaluation of short-, medium-, and long-run fiscal implications of alternative reform strategies, including administrative restructuring, pension reform, and the gradual reallocation of resources toward human capital and productivity-enhancing investment. In contrast to descriptive expenditure reviews based on static budget comparisons, the model explicitly incorporates nonlinear transition dynamics and institutional frictions. Simulations show that structural expenditure reforms generate significant transitional fiscal costs arising from overlapping institutional systems, labor adjustment frictions, and pension transition liabilities. As a result, fiscal reform produces a J-shaped expenditure trajectory in which total spending initially increases before gradually converging toward a more efficient long-run allocation. These findings highlight the importance of accounting for adjustment costs and transition dynamics when evaluating the feasibility and timing of structural fiscal reforms.

Task-based models of AI and labor hold organizational structure fixed. We introduce agent capital: AI that reduces coordination costs, expanding spans of control and enabling endogenous task creation. Five propositions characterize how coordination compression affects output, hierarchy, manager demand, wage dispersion, and the task frontier. The model generates a regime fork: the same technology produces broad-based gains or superstar concentration depending on who benefits from coordination compression. Simulations with heterogeneous workers confirm sharp regime divergence. Economy-wide inequality falls in all regimes through employment expansion, but the manager-worker wage gap widens universally. The distributional impact hinges on who controls organizational elasticity.

Multivariate shortfall risk measures provide a principled framework for quantifying systemic risk and determining capital allocations prior to aggregation in interconnected financial systems. Despite their well established theoretical properties, the numerical estimation of multivariate shortfall risk and the corresponding optimal allocations remains computationally challenging, as existing Monte Carlo based approaches can be numerically expensive due to slow convergence. In this work, we develop a new class of single and multilevel numerical algorithms for estimating multivariate shortfall risk and the associated optimal allocations, based on a combination of Fourier inversion techniques and randomized quasi Monte Carlo (RQMC) sampling. Rather than operating in physical space, our approach evaluates the relevant expectations appearing in the risk constraint and its optimization in the frequency domain, where the integrands exhibit enhanced smoothness properties that are well suited for RQMC integration. We establish a rigorous mathematical framework for the resulting Fourier RQMC estimators, including convergence analysis and computational complexity bounds. Beyond the single level method, we introduce a multilevel RQMC scheme that exploits the geometric convergence of the underlying deterministic optimization algorithm to reduce computational cost while preserving accuracy. Numerical experiments demonstrate that the proposed Fourier RQMC methods outperform sample average approximation and stochastic optimization benchmarks in terms of accuracy and computational cost across a range of models for the risk factors and loss structures. Consistent with the theoretical analysis, these results demonstrate improved asymptotic convergence and complexity rates relative to the benchmark methods, with additional savings achieved through the proposed multilevel RQMC construction.

Stablecoins have historically depegged due from par to large sales, possibly of speculative nature, or poor reserve asset quality. Using a global game which addresses both concerns, we show that the selling pressure on stablecoin holders increases in the presence of a large sale. While precise public knowledge reduces (increases) the probability of a run when fundamentals are strong (weak), interestingly, more precise private signals increase (reduce) the probability of a run when fundamentals are strong (weak), potentially explaining the stability of opaque stablecoins. The total run probability can be decomposed into components representing risks from large sales and poor collateral. By analyzing how these risk components vary with respect to information uncertainty and fundamentals, we can split the fundamental space into regions based on the type of risk a stablecoin issuer is more prone to. We suggest testable implications and connect our model's implications to real-world applications, including depegging events and the no-questions-asked property of money.

We conduct experiments with algorithmic pricing agents based on Large Language Models (LLMs). In oligopoly settings, LLM-based pricing agents quickly and autonomously reach supracompetitive prices and profits. Variation in seemingly innocuous phrases in LLM instructions ("prompts") substantially influence the degree of supracompetitive pricing. We develop novel techniques for behavioral analysis of LLMs and use them to uncover price-war concerns as a contributing factor. Our results extend to auction settings. Our findings uncover unique challenges to any future regulation of LLM-based pricing agents, and AI-based pricing agents more broadly.

Through the reinterpretation of housing data as candlesticks, we extend Nature Scientific Reports' article by Liang and Unwin [LU22] on stock market indicators for COVID-19 data, and utilize some of the most prominent technical indicators from the stock market to estimate future changes in the housing market, comparing the findings to those one would obtain from studying real estate ETF's. By providing an analysis of MACD, RSI, and Candlestick indicators (Bullish Engulfing, Bearish Engulfing, Hanging Man, and Hammer), we exhibit their statistical significance in making predictions for USA data sets (using Zillow Housing data) and also consider their applications within three different scenarios: a stable housing market, a volatile housing market, and a saturated market. In particular, we show that bearish indicators have a much higher statistical significance then bullish indicators, and we further illustrate how in less stable or more populated countries, bearish trends are only slightly more statistically present compared to bullish trends.