From the 1980s to the early 2000s, Japan experienced a real estate bubble followed by a prolonged period of economic stagnation known as the "lost decade". This study examines the dynamic relationships among monetary liquidity, bank credit, corporate leverage, asset prices and real economic activity in Japan between 1980 and 2000. Using quarterly data from official sources covering 1980Q1 to 2000Q4, it estimates a six-variable vector autoregression (VAR) comprising M2, bank credit, corporate leverage, a real estate price index, the Nikkei 225 and real gross domestic product (GDP). Together, the variables represent a possible transmission sequence from monetary liquidity and bank credit to asset prices, corporate balance sheets and real economic activity. The estimated equations differ markedly in explanatory power. The real estate price equation has the strongest in-sample fit, with an adjusted R² of 0.672, and the first lag of M2 is positive and statistically significant at the 1% level in that equation. A positive change in real estate price growth is associated with a higher change in GDP growth in the following quarter, and a positive change in corporate leverage is associated with a higher change in bank credit growth two quarters later. By contrast, the M2 equation is not jointly significant, and the system does not explain changes in M2 well. Unit-root tests support first-differencing; the M2–real estate and real estate–GDP relationships persist with three lags, but the leverage–credit relationships weaken. These relationships are reported as dynamic associations rather than as evidence of causation.
Gold is generally regarded as a safe-haven asset, but its price behavior differs in the different phases of a crisis. This paper investigates the co-movement of gold prices with economic policy uncertainty, market risk sentiment, and key financial market variables using monthly data from January 2006 to December 2025 and compares gold price performance during the 2008 global financial crisis and the 2020 COVID-19 pandemic. Descriptive statistics, trend comparison and case analysis are used to investigate the variations of the Global Economic Policy Uncertainty Index (GEPU), VIX, the US Dollar Index, the US 10-year real interest rate and the S&P 500 Index. The analysis reveals that gold prices do not always move together with risk indicators. Gold might be under some pressure at first as the US dollar strengthens and liquidity pressures build up in the early stages of a crisis, but gold prices may rise again as real interest rates fall and the financial market environment changes. Gold's function as a safe-haven asset was phased and conditional in both crises.
This paper investigates how city level economic and demographic conditions shape firm-survival across Chinese cities. Using 2000-2024 panel data covering 50 cities and over 5 million firms, this study applies principal-component analysis (PCA) to handle high indicator correlation and constructs three composite indices: economic opportunity, business-pressure, and population-base index. These indices are incorporated into weighted least-squares regressions. The baseline specification controls for year, industry and firm ownership fixed effects, while the preferred model further adds city fixed effects; unweighted OLS serves for robustness checks. Baseline results show the population-base index has a significantly negative association with firm-survival, whereas economic opportunity and business-pressure indices are statistically insignificant at conventional levels. Once city fixed effects are included, all three indices lose significance. This suggests observed cross-city firm-survival gaps stem largely from persistent intercity structural differences rather than short-run within-city changes. Robustness tests reveal operating pressure correlates negatively with firm-survival rates. Overall, firm-survival disparities are pronounced across Chinese cities, requiring researchers to distinguish cross sectional city differences from over time within-city variations when analysing urban environment firm-survival linkages.
With AI agents increasingly involved in teamwork, whether AI can substitute human collaboration has become a heated topic in academia. Large-model-driven agents are now capable of participating in team brainstorming, task coordination, and collaborative problem-solving, and knowledge-intensive industries are beginning to form a large number of human-machine hybrid teams. Existing research mostly relies on theoretical reasoning and lacks empirical perceptual data from front-line practitioners. Drawing on a questionnaire survey of 158 employees with cross-industry teamwork experience, this paper explores practitioners' perceptions regarding the substitution of human collaborative behaviors by AI. The results show that AI can only partially substitute procedural and rule-oriented collaborative tasks, whereas social-emotional communication, conflict mediation and value judgment remain irreplaceable human strengths. Position, work experience and AI familiarity jointly shape individual perceptions of the substitution boundary. Most practitioners regard AI as an auxiliary tool rather than a leader of team collaboration. This study provides a practical reference for enterprises applying AI collaboration tools.
