This study shows that technological specialization is one plausible driver of long-term international capital flows. The underlying mechanism comprises the two sides of capital scarcity: although a shortage of capital currently generates a higher marginal product, it also makes a country inclined toward capital-saving technology. Therefore, in equilibrium, the rate of capital returns is jointly determined by a convergence effect from the law of diminishing return and an opposing divergence effect from directed technological change. Initially capital-rich countries favor capital-complements-biased technology and continuously import capital from capital-poor countries that develop capital-substitutes-biased technology. We name this anti-convergence force on capital allocation the capital Matthew effect. This new perspective is consistent with many related international finance puzzles, including the Lucas paradox, global imbalance, and allocation puzzle. Finally, we provide both country-level and firm-level evidence supporting this new model mechanism.

Capital inflow surges destabilize the economy through a maturity shortening mechanism. The underlying reason is that firms have incentives to redeem their debt on demand to accommodate the potential liquidity needs of global investors, which makes international borrowing endogenously fragile. Based on a theoretical model and empirical evidence at both the firm and macro levels, our main findings are twofold. First, a significant association exists between surges and shortened corporate debt maturity, especially for firms with foreign bank relationships and higher redeployability. Second, the probability of a crisis following surges with a flattened yield curve is significantly higher than that following surges without one. Our study suggests that debt maturity is the key to understand the financial instability consequences of capital inflow bonanzas.

This study provides firm-level evidence on the effect of capital account liberalization on total factor productivity (TFP) growth. We find that a one standard deviation increase in the capital account openness indicator constructed by Fernández et al. (2016) is significantly associated with a 0.18 standard deviation increase in firms’ TFP growth rates. The productivity-enhancing effects are stronger for sectors with higher external finance dependence and capital-skill complementarity, and are persistent five years after liberalization. Moreover, we show that potential transmission mechanisms include improved financing conditions, greater skilled labor utilization, and technology upgrades. Finally, we document heterogeneous effects across firm size and tradability, and threshold effects with respect to the country's institutional quality.

Eighty percent of US corporate debt is cash-flow-based, yet most macroeconomic models rely on collateral constraints. We build a continuous-time model with two coupled distributions of heterogeneous firms, one for each constraint type, where collateral-based and earnings-based borrowing constraints coexist, and establish three contributions. First, the earnings-based leverage multiplier is structurally convex in productivity, activating a Jensen channel through which micro uncertainty shocks expand aggregate credit while simultaneously elevating systemic fragility, a mechanism absent in collateral economies. Second, earnings-based leverage depends on the equilibrium wage through the profit rate, unlike collateral-based leverage which is wage-insensitive. This creates a pecuniary externality through wages: collective borrowing bids up wages, compressing profit margins and tightening earnings constraints for all cash-flow borrowers, an effect each firm individually ignores. Third, we construct an Earnings-Based Stress Index (EBSI), defined as the average leverage utilization of cash-flow-constrained firms, which measures how close the sector is to its aggregate borrowing limit. Because the EBSI tracks both systemic fragility and the magnitude of the wage externality, it serves as a sufficient statistic for optimal policy: the optimal tax on earnings-based debt is 2.15 times the tax on collateral-based debt, and a simple EBSI-indexed rule captures 72% of the welfare gain from the fully optimal policy.

Rising market concentration need not signal declining competition. We show that intangible capital accumulation endogenously generates a sigmoidal production function featuring a Valley of Death---an increasing-returns region firms must traverse before becoming profitable. As economies become more intangible-intensive, this Valley expands, simultaneously raising the share of unprofitable firms and market concentration under perfect competition. We construct an Efficient Concentration Frontier that decomposes observed HHI into technological and market-power components. Bayesian MCMC estimation across the U.S. and eleven countries confirms the predicted shift to sigmoidal production since the 1980s. In a calibrated Hopenhayn model, welfare-relevant excess concentration is consistently negative: the entire observed rise in HHI reflects technology, not market power.

What matters for intangible-intensive production is not credit volume but composition---the mix of cash-flow-based versus asset-based lending. Using data from 27 countries and over 160,000 firm-year observations, we show that cash-flow-based lending drives intangible investment while overall credit depth does not, and that social trust is its deep institutional determinant. A firm-level trust measure from LLM analysis of 700,000 earnings call transcripts confirms that more transparent firms obtain more cash-flow debt. Two symmetric causal tests sharpen identification: insolvency reforms disproportionately benefit low-trust firms, while the financial crisis disproportionately hurts high-trust firms whose CF lending channel is disrupted. A model with strategic complementarity between financial and production structure generates multiple equilibria: high-trust economies reach an intangible frontier while low-trust economies remain trapped in tangible production. Eliminating cash-flow frictions raises TFP by 33 percent; composition policies generate two to three times larger gains than volume policies.

Using patent-level data matched to firm characteristics from 1962--2023, we document that during industry distress the market undervalues patents by approximately 37% relative to fundamental value, a negative bubble in the market for innovation. Standard no-bubble theorems rule out negative bubbles under limited liability and transversality, and the classic Miller (1977) mechanism predicts that short-sale constraints push prices up, not down. We propose a resolution through a dynamic trading model with three progressively reinforcing mechanisms. First, the confident pessimist channel creates the negative bubble: when bearish investors hold their views with greater precision than optimists, the precision-weighted consensus price falls below every investor's individual valuation. Analyst forecast data confirm that the normally positive optimism-confidence correlation reverses during distress. Second, capital-constrained forced selling amplifies the discount: binding leverage constraints force informed optimists to liquidate, triggering a self-reinforcing fire-sale loop. Third, the repurchase option, the symmetric counterpart of the Harrison-Kreps resale option, sustains the underpricing as unconstrained optimists rationally delay purchase in anticipation of further price declines. Micro-level 13F data validate the second and third channels: nearly half of incumbent active investors exit during distress while an equal number of new active investors enter, confirming the model's partition of optimists into forced sellers and dry-powder entrants. A gradual resolution of negative bubbles is observed empirically, consistent with the significant value-weighted alpha of 7.6% per year earned by distress-period innovation portfolios.

China's infrastructure bias, overcapacity, and structural trade surplus are commonly attributed to political tournament incentives. We offer a fiscal explanation. Under China's origin-based tax sharing, local governments retain 50% of VAT at the point of production but receive no consumption tax revenue. This fiscal wedge makes production a fiscal asset and consumption a liability, generating infrastructure overinvestment and welfare suppression even under benevolent governance. We validate the mechanism with 540,000 industrial land transactions: cities hit harder by the 2016 Business-to-VAT reform compress land prices to attract manufacturing, with effects concentrated among younger officials and high-VAT industries. Business cycle accounting across 283 prefectures reveals that fiscal forces explain 78% of post-2016 investment dynamics, up from 36% before 2008. Policy counterfactuals show that destination-based tax reform raises consumption by 8.8 percentage points and delivers welfare gains four times larger than equivalent transfers.

We develop a quantitative Stablecoin--Intermediary New Keynesian (S-INK) framework in which stablecoin issuers function as intermediaries supplying dollar-denominated digital liquidity. We empirically document that monetary policy shocks in either direction---both tightening and easing---are followed by significant stablecoin supply contraction and elevated market-implied depeg probability. These empirical patterns motivate and discipline a structural model. Unlike traditional banks, issuers fund themselves with liquid, zero-interest liabilities yet must earn returns by investing in duration-bearing assets, endogenously exposing their balance sheets to interest rate risk. The model implies a narrow corridor for monetary policy, bounded by the valuation wall at high rates---where mark-to-market losses erode net worth and trigger strategic depeg---and the instability trap at low rates---where compressed seigniorage produces slow-bleed insolvency. Our baseline calibration generates a partial equilibrium corridor of 453 basis points; general equilibrium feedback through price stickiness compresses this to approximately 217 basis points. The corridor closes as stablecoin adoption grows, requiring a factor of 3.5 increase in adoption in partial equilibrium or 50 percent in general equilibrium. Our S-INK framework establishes a regulatory trilemma: par convertibility, adequate seigniorage, and monetary sovereignty cannot be jointly achieved.

This paper investigates how BigTech lenders and traditional banks differentially respond to monetary policy changes. We posit that BigTech lenders possess an information advantage that diminishes with firm size, resulting in lower lending ambiguity when serving smaller firms relative to banks. Using a granular micro-level dataset on small-business loans from both types of lenders, we show that BigTech lenders are significantly more responsive to monetary policy on the extensive margin, while differences on the intensive margin are statistically insignificant. Moreover, BigTech lenders react more strongly during periods of monetary easing than tightening, indicating an asymmetric transmission pattern. We further examine adjustments in other loan terms, including interest rates and maturity, and discuss how regulatory interventions targeting BigTech credit may shape these dynamics.

We study an incentive-compatible mechanism-embedding financial incentives into non-financial actions-that fosters individual environmental engagement and facilitates the private sector's internalization of climate externalities. Using a novel dataset of 100,000 randomly selected users from Ant Forest, a widely used personal carbon tracking program within Alipay-China's leading BigTech platform, we demonstrate that tying eco-friendly behaviors to credit limit adjustments encourages users to engage in green actions. The platform benefits from reduced default risk even amid credit expansion, likely driven by a signaling mechanism in which costly green actions reveal environmental type. Climate-responsible individuals often exhibit conscientious and disciplined behavior across various domains, allowing lenders to infer creditworthiness from green actions. Our structural model estimates an annual green value of $413.20 million generated by linking credit access to green actions. This incentive-based approach yields larger welfare gains than traditional policy instruments such as mandates or subsidies, particularly when public green awareness is low. Our findings identify the screening role of green behaviors in household lending to align environmental values with financial value and highlight alternative data as a viable source for credit allocation.

This paper shows that increasing returns to scale steepen the earnings-productivity gradient, redistributing both income and risk toward more productive firms. In the presence of financing frictions, the negative impact of heightened risk dominates the positive effects of increased income. As a result, more productive firms accumulate larger cash reserves and reduce investment further below their first-best levels, exacerbating capital misallocation. The calibrated model replicates key empirical patterns, including the rise in cash reserves and the decline in capital allocation efficiency among U.S. public firms.

This paper examines the real effects of firm-level uncertainty amid financial friction. Empirically, we show that productivity explains 68% of the negative impact of firm-level uncertainty on output, while reduced physical investment accounts for only 3%. Using a dynamic model with endogenous productivity growth and defaultable debt, we show that uncertainty tightens borrowing limits by raising default risk. Unlike aggregate uncertainty, firm-level uncertainty has little effect on asset reallocation value, leading firms to favor physical capital accumulation over productivity-enhancing efforts to support liquidation value and debt pricing.

This paper is the first to study the effects of capital account liberalization on structural transformation and compare the contribution of within term and structural term to economic growth. We use a 10-sector-level productivity data set to decompose the effects of opening capital account on within-sector productivity growth and cross-sector structural transformation. We find that opening capital account is associated with labor productivity and employment share increment in sectors with higher human capital intensity and external financial dependence, as well as non-trad-able sectors. But it results in a growth-reducing structural transformation by directing labor into sectors with lower productivity. Moreover, in the ten years after capital account liberalization, the contribution share of structural transformation decreases while that of within productivity growth increases. We conclude that the relationship between capital account liberalization and economic growth is within gain and structural pain.

We develop a theoretical model to explain the shift in risk transmission between the dollar and crypto markets, focusing on stablecoins while excluding other channels like institutional investors. Our model shows that risk flows from the dollar market into the crypto market as stablecoin supply expands, and flows into the dollar market when stablecoin reserves contract. This transmission mechanism can be asymmetric when stablecoins hold sufficient reserves to absorb crypto-market shocks. Copula-based CoVaR tests support our model. Akin to securitization during the 2007-08 financial crisis, stablecoins’ role as a bridge highlights how niche markets can amplify systemic risks.