Biodiversity risk is increasingly becoming a corporate sustainability concern that extends beyond a firm’s operations to its supply chains. Using supplier-customer data from 2009 to 2023, we explore how supplier biodiversity risk influences the environmental, social, and governance (ESG) behavior of downstream firms (customers). Notably, supplier biodiversity risk significantly improves the ESG performance of downstream firms; however, this improvement is superficial and, in reality, the result of greenwashing. Mechanism analysis shows that this phenomenon is primarily driven by three paths: supply-chain stickiness, reputational pressure, and capital flow risk. Furthermore, downstream firms with lower levels of supply-chain finance, higher supply-chain operational risk, weaker supply-chain resilience, weaker internal controls, and a higher shareholding ratio of institutional investors are more significantly affected by supplier biodiversity risk. Overall, our findings provide novel evidence that biodiversity risk is transmitted through supplier-customer relationships and reveal a previously undocumented consequence of such transmission: observable ESG improvement may conceal defensive greenwashing rather than substantive sustainability progress.
This paper provides new, structural estimates of Okun’s unemployment-output relationship for euro area countries between 1979 and 2019. We show that these structural estimates are stable over time and yet substantially smaller than the reduced-form estimates that tend to characterise the literature. We also find that country specific factors largely shape how output responds to unemployment in both core and periphery economies. Specifically, for the euro periphery we find that product market regulation plays a major role in explaining the significance of Okun’s estimates. Our results are robust, inter alia, to conditioning on diverse institutional set-ups.
Analyzing more than 4.3 million job postings by Chinese listed firms from 2015 to 2021, we show that firms with higher green hiring experience significantly lower environmental, social, and governance (ESG) rating disagreement. Green job postings function as credible, albeit costly, genuine ESG engagement signals, improving the quality of the common information set available to ESG raters and reducing their reliance on subjective assessments. This is consistent with a separating equilibrium: the disagreement-reducing effect is stronger when there are higher net benefits of green hiring and potential penalties for deceptive signaling. Rating agencies subsequently validate the signal, as green-hiring firms exhibit stronger ex post ESG performance and provide more objective ESG disclosures. Moreover, investors, particularly foreign ones, respond positively to green hiring. Further, reduced ESG-related uncertainty is associated with a lower equity risk premium, greater overseas expansion, and improved market efficiency. Overall, green hiring serves as an effective labor-market-based signal that mitigates information frictions in ESG evaluation, thereby facilitating more efficient cross-border capital flows and supporting firms’ global expansion under severe information asymmetry.
This paper explores the effect of the euro on business cycle synchronization (BCS) of Economic and Monetary Union (EMU) countries. Euro-area countries with a longer exposure to the common currency experienced lower BCS in the post-EMU period. Output co-movement declined unevenly more in EMU countries compared to non-euro area economies. Mechanism analysis highlights the decline of the tradable goods’ sector more sharply in the EMU than in the rest of the world as α possible driver of this result. Heavily regulated EMU countries experienced lower synchronization in the post euro period. These results are confirmed after accounting for a rich set of confounding variables, trends and shocks.
Can macro-financial conditions amplify adverse shocks originating in the banking system? We argue that when vulnerability in the banking system increases, the extent to which financial intermediation and the macroeconomy are adversely affected is likely to depend on whether banks are operating in normal times or under more challenging conditions. Using a Bayesian Panel Threshold VAR, data on euro area countries and a high-frequency, forward-looking measure of the banking system’s vulnerability to market-based equity losses, we show that nonlinearities matter. When the banking system is already in distress, lending and economic activity contract more severely following a further increase in banks’ vulnerability. Similarly, low interest rates, typically associated with low bank margins and profitability, correspond to larger declines in lending. In both environments, corporate lending contracts more than mortgage lending, a finding consistent with de-risking. The state of the business cycle, though, does not alter shock propagation to the same extent. Our findings suggest that it is particularly important for macroprudential policies to preserve banks’ resilience in times of stress, characterized by elevated banking system vulnerability and low interest rates.
This paper reassesses the debate on the end of the Chinese silver standard (CSS) in the mid 1930s. One side claims the U.S. Silver Purchase Act of June 1934 drained China of silver, which produced deflation and economic crises that destabilized the CSS. Others argue the CSS collapsed because its operating mechanism was inherently unstable. We look for evidence of instability in the CSS by estimating Bayesian structural VARs with drifting parameters on new China-U.K. and China-U.S. samples from April 1912 to September 1934. Our estimates show uncovered interest parity and long-run arbitrage often held under the CSS while its instability peaked during the U.K. and U.S. recessions of the early 1920s and in 1931 during the Great Depression. We argue these results show the demise of the CSS was due to reasons other than the U.S. Silver Purchase Act of June 1934 or its own design flaw.
The rapid expansion of cross-border capital flows has emerged as a defining feature of the world economy in recent decades. This expansion is often attributed to the wave of capital account liberalization that spread during the late 20th century and reached its height just before the global financial crisis (GFC). The GFC exposed the risks associated with unregulated capital flows and, in turn, revived interest in the use of capital controls. But are capital controls effective in curbing cross-border flows? We present an updated and extended dataset on capital controls with near-global coverage, building on Fernandez et al. (2016). The dataset provides rich granularity, offering disaggregated indices by transaction type, asset class, direction of flow, and residency for 179 countries from 1999 to 2022. We draw on this database to document stylized facts on the use of capital controls across countries and over time. We also use this novel dataset to study the effectiveness of capital controls in curbing cross-border flows, finding limited aggregate effects but pronounced heterogeneity across instruments, flow direction, and whether measures target residents or nonresidents. A central finding of the paper is that restrictions targeting nonresidents are the most effective in curbing capital inflows. Strikingly, restrictions on outflows targeting nonresidents exert a strong and robust negative effect on foreign capital inflows. By contrast, restrictions targeting residents are largely ineffective.
We examine whether cultural trust biases exist in international syndicated loans and find that the more positive the perception of trustworthiness that the lender's country has for the borrower's country, the lower loan spreads the lender will charge the borrower. The results remain consistent across robustness checks, including alternative measures of cultural trust bias, instrumentalvariable regression, and the inclusion of important loan contract determinants such as bank relation, geographical distance, cultural distance, expected default frequency, and creditworthiness. We use several short-term shocks to the perception of trustworthiness to address potential endogeneity concerns, namely (i) British bias towards the French during the Iraq war in 2003, (ii) North-South European biases during the European debt crisis, and (iii) the US Presidential Election of 2016. Consistent with our trust bias interpretation, the results are stronger for borrowing firms that bear their country's name and for the top quartile of borrowers from countries perceived as the most trustworthy.
The spanning hypothesis in government bond markets posits that a small number of yield-curve factors fully summarize the information relevant for pricing and forecasting bond returns. Despite extensive empirical investigation, evidence on the validity of this hypothesis remains mixed. Concurrently, recent advances in machine learning have renewed interest in uncovering nonlinear predictive relationships between macroeconomic conditions and the term structure of interest rates. This paper provides a systematic literature review of studies that employ machine learning methods to forecast bond yields and excess returns, with a particular focus on their implications for the spanning hypothesis. We synthesize evidence across datasets, model classes, validation strategies, and approaches to economic interpretation. The literature is highly fragmented, lacking a coherent model hierarchy or standardized out-of-sample evaluation framework. Nonetheless, evidence from methodologically rigorous studies suggests that nonlinear methods, in particular parsimonious neural networks and tree-based ensembles, can extract predictive signals from macroeconomic data that are not spanned by conventional yield-curve factors. At the same time, we document that reported forecasting gains are frequently overstated due to methodological shortcomings, including inappropriate evaluation metrics, look-ahead bias arising from revised macroeconomic data, and data leakage induced by global pre-processing and model selection. Taken together, our findings imply that while the true predictive advantage of machine learning over traditional term-structure benchmarks is more modest than often claimed, it remains economically meaningful.
China has established the largest carbon emission trading system (CETS) among developing countries, a market-based environmental scheme, to address environmental challenges. However, the effect of the CETS on firms’ green development remains a question. This paper investigates the impact of the CETS on firms’ green development from a perspective of market mechanisms. Using a novel dataset identifying compliance entities, this study constructs a staggered difference-in-differences (DID) model using the establishment of pilot carbon markets as a natural experiment. The findings indicate that the CETS significantly facilitates firms’ green development. A series of additional tests confirms the causality. Further analysis reveals that higher carbon prices stimulate green development, confirming the price signal of carbon prices in guiding firms’ behaviors. Cross-sectional analysis reveals that mature market mechanisms, including quota accounting methods, quota allocations, and market liquidity, enhance the CETS’s environmental effect. Economic impact shows that firms emphasize green innovation and enhance profitability in the long run. Overall, this study provides suggestive evidence consistence with market-oriented channels, and offers important implications for advancing carbon trading markets globally.
Geoeconomic fragmentation—the phenomenon of international transactions being increasingly restricted to politically aligned partners—creates risks for individual countries but also opportunities that some hope to seize by becoming “connector” countries. We modify a standard trade model by introducing iceberg costs that increase with geopolitical distance between country pairs. The response of per capita consumption to a geopolitical shock is shown to depend on two related but distinct indices: vulnerability, which is a country’s transaction-weighted geopolitical distance from its trade partners, and connectedness, which is a country’s transaction-weighted standard deviation of geopolitical distance from trade partners. The latter captures a country’s geopolitical diversification. We distinguish between this type of connectedness and the supply chain-related connectedness discussed by previous studies, arguing that the horizontal measure is more relevant in a geoeconomically fragmenting world. We construct a comprehensive database to examine geoeconomic vulnerability and connectedness across multiple types of international transactions, documenting several stylized facts.
Natural disasters often have high economic costs, setting back years of investment in developing countries. This paper develops a multi-sector DSGE model to study the macroeconomic and welfare implications of financing resilience-building using different fiscal instruments. The model includes developing countries’ macroeconomic and distributional features, such as a large unproductive rural sector, an incomplete credit market, and an informal sector. The results show that investing in resilience capital in a disaster-prone country improves welfare despite its high economic costs, but the financial instrument used to mobilize revenue matters. The benefits of resilience building are disproportionately concentrated among wealthier households employed in capital-intensive sectors, while rural and unskilled workers — who lack access to credit markets and work in sectors that do not directly employ public capital — experience smaller gains. Income inequality, as measured by the Gini index, increases slightly with resilience investment, suggesting a role for complementary redistributive policies.
We show that a Bayesian mean-variance (MV) optimisation can substantially enhance the performance of established currency factor strategies such as carry, value, and momentum. We find that the improved performance is due to the stronger cross-sectional predictability of the optimised strategies, instead of the time series predictability. International diversification plays an important role in the outperformance of the MV optimisation approach. Our asset pricing tests suggest that the MV-optimised factors subsume the abnormal return of the naïve currency factors.
To theoretically investigate the mechanism of stablecoin fragility under information shocks, we construct a novel agent-based market empowered by Large Language Models (LLMs). This framework reproduces the micro-cognitive dynamics of investor panic, allowing agents to process unstructured narrative data. We find that stablecoin collapse is a highly non-linear phenomenon governed by a critical severity threshold. Specifically, while the peg remains robust under moderate stress, crossing the narrative severity threshold triggers a "cognitive de-pegging": the expected maximum price deviation surges by approximately 1,441 basis points as diverse investor beliefs rapidly converge into synchronized selling. This highlights that the intensity of a narrative, rather than its specific content, is the primary driver of digital bank runs, suggesting that regulators should implement "narrative stress tests" to monitor systemic risk.
This paper provides the first systematic international examination of option-implied lower bounds on the expected market risk premium (EMRP). Using the Chabi-Yo and Loudis (2020) bound as our primary measure, we apply the Back et al. (2022) validity and tightness tests to 15 international stock markets and find that the bound is both valid and tight at the one-month horizon in 12 of these markets, supporting its use as an unbiased proxy for expected returns. Using these validated forward-looking measures, we revisit the international asset pricing literature and show that exposure to global market, regional market, and global financial cycle risk is priced, while exposure to dollar risk, carry risk, and the non-market factors in the Fama and French (2017) global five-factor model are not. Notably, several of these relationships cannot be identified using realized returns over the same sample period, illustrating the value of direct measures of expected returns in asset pricing tests. Finally, we show that option-implied EMRPs significantly predict the cross-section of realized international stock market returns. Our results are robust to using the alternative lower bound of Martin (2017).
This paper provides empirical evidence on the balance sheet channel of fiscal policy in peripheral European economies. Our findings using a Panel VAR, reveal that shifts in financial institutions' balance sheets following a debt-financed fiscal expansion, reduce credit provision and investment in these countries. Moreover, the analysis indicates that economies with higher sovereign exposure experienced higher credit crunches and investment declines. To explore the underlying mechanisms, we estimate a DSGE model that incorporates banks as primary holders of sovereign debt. The model shows that sovereign exposure amplifies the negative effects on credit supply, lowering investment and capital formation. A counterfactual scenario without bank-held sovereign bonds isolates the contribution of the balance sheet channel: removing this channel weakens the crowding-out effect, with investment falling 0.2 percentage points less and output increasing by 0.02 percentage points more. These effects appear stronger during the financial and sovereign debt crises.
We investigate nonlinear real exchange rate (RER) dynamics using a high-frequency panel dataset of Apple iPad prices across global markets. Our findings challenge the conventional view of slow RER adjustment by demonstrating that larger mispricings are corrected much faster, while smaller deviations persist for longer periods. We identify a valuation threshold effect, where smaller deviations behave like unit root processes, while larger discrepancies are more quickly mean-reverting. Furthermore, the magnitude of these thresholds decreases as the evaluation time horizon increases. Transaction costs play a crucial role in shaping these dynamics, with higher costs associated with wider thresholds for arbitrage activity. Additionally, the results reveal asymmetry in RER adjustments, where undervaluations are corrected more rapidly than overvaluations. These findings provide new insights into the speed and nature of RER adjustments, suggesting that product-specific price data, such as that from iPads, can offer a more granular understanding of international price convergence.
We investigate the rationale for the existence of the Zero Lower Bound (ZLB) and the role of Central Bank Digital Currency (CBDC) in combating the ZLB. The ZLB arises due to the fixed nominal interest rate on cash and the central bank's desire to prevent financial disintermediation. The impact of CBDC on the ZLB depends on the existence of cash. If CBDC is introduced to completely replace cash, it can help breach the ZLB. However, if cash remains in circulation alongside deposits and CBDC, the coexistence of these currencies would exacerbate the ZLB problem through intensified currency competition.
