G20 资源重配置效率（RE）指数
The G20 Resource Reallocation Efficiency (RE) Index
本页汇总一项独立的实证研究成果：基于公开、可核验的国际数据源，对 18 个 G20 经济体在 2000–2023 年的资源重配置效率进行量化测算，构建 V（反应弹性）、C（转化率）、F（制度摩擦）三维分数及 RE = VC/(1+F) 综合指数。数据表格、排行榜与图表均由下方真实面板数据实时计算生成，而非预设结论。
This page presents an independent empirical research deliverable: using public, verifiable international data sources, it quantifies resource-reallocation efficiency for 18 G20 economies over 2000–2023, constructing V (Responsive Elasticity), C (Conversion Rate), and F (Institutional Friction) sub-scores plus the composite index RE = VC/(1+F). All tables, rankings, and charts below are computed live from the underlying real panel data, not asserted in prose.
从理论框架到可观测指数
From Theoretical Framework to an Observable Index
三维操作化
Three-Dimensional Operationalization
V（反应弹性）以财政支出结构对外部冲击的调整速度和幅度代理；C（转化率）以技术、资本与政策产出的转化效率代理；F（制度摩擦）以官僚质量、监管负担与政治摩擦的反向合成代理。三者经 Min-Max 标准化后，按 RE = VC/(1+F) 聚合为 0–100 区间的综合指数。
V (Responsive Elasticity) proxies the speed and magnitude of fiscal-structure adjustment to external shocks; C (Conversion Rate) proxies the efficiency with which technology, capital, and policy inputs convert into outputs; F (Institutional Friction) proxies bureaucratic quality, regulatory burden, and political friction (reverse-coded). After Min-Max normalization, the three are aggregated via RE = VC/(1+F) into a composite index rescaled to 0–100.
数据来源清单
Data Source Inventory
- 世界银行《世界发展指标》(World Bank WDI) — 宏观经济与财政结构变量
- World Bank World Development Indicators — macroeconomic & fiscal-structure variables
- Quality of Government（QoG）标准化数据集 — 治理质量与制度变量
- Quality of Government (QoG) Standard Dataset — governance & institutional variables
- V-Dem 民主多样性项目 — 政治问责与国家能力指标
- V-Dem (Varieties of Democracy) — political accountability & state-capacity indicators
- Fraser Institute 经济自由度指数（EFW） — 监管负担与市场自由度
- Fraser Institute Economic Freedom of the World (EFW) — regulatory burden & market freedom
- ICRG 国别风险指南 — 政治稳定性与制度风险
- ICRG Country Risk Guide — political stability & institutional risk
构建方法
Construction Methodology
对缺失值采用迭代式多重插补（MICE 思路）处理；各分维度权重通过五年滚动窗口的主成分分析（PCA）动态估计，以捕捉制度重要性随时间的变化；综合指数以乘法比率形式聚合（RE = VC/(1+F)），并通过 500 次自举（Bootstrap）重采样估计置信区间与标准误，量化测算的不确定性。
Missing values are handled via iterative multiple imputation (a MICE-style approach); sub-dimension weights are estimated dynamically via Principal Component Analysis (PCA) over rolling five-year windows to capture how institutional importance shifts over time; the composite is aggregated as a multiplicative ratio (RE = VC/(1+F)), with 500-replicate Bootstrap resampling used to estimate confidence intervals and standard errors, quantifying measurement uncertainty.
可靠性与局限性并重
Reliability Alongside Limitations
该指数基于真实、可公开核验的国际数据构建，方法论在统计上稳健；但仍存在跨国可比性、代理变量效度、插补依赖假设等固有局限，详见下文"稳健性与局限性"一节。本页呈现的所有数值均为研究性估计，供学术讨论与理论验证参考，不构成官方统计发布。
The index is built from real, publicly verifiable international data with a statistically sound methodology; however, it retains inherent limitations around cross-national comparability, proxy-variable validity, and imputation assumptions — detailed in "Robustness & Limitations" below. All figures on this page are research estimates for academic discussion and theory-testing, not an official statistical release.
面板数据表：18 个 G20 经济体，2000–2023
Panel Data Table: 18 G20 Economies, 2000–2023
年份选择器控制主表格与排行榜展示的截面（2000–2023 共 24 个年份）；下方"时间序列趋势"图始终展示已选国家在全部 24 年中的完整变化路径。
The year selector controls the cross-sectional snapshot shown in the main table and leaderboard (24 years, 2000–2023); the "Time-Series Trend" chart below always shows the full 24-year trajectory of the selected countries.
| 国家 | Country | V | V | C | C | F | F | RE | RE | 操作 | Action |
|---|
RE 综合得分排行榜（当前筛选年份）
RE Composite Score Ranking (Current Filtered Year)
随上方年份选择器自动更新，反映当前筛选区域内各国的相对排名（18 国样本，不含中国）。
Updates automatically with the year selector above, reflecting relative rankings among filtered countries (18-country sample, excluding China).
2000–2023 RE 变化趋势
RE Trajectory, 2000–2023
此图不受上方年份选择器影响，始终展示已选国家（或默认当前年份 Top 6）在 2000–2023 全部 24 个年份中的 RE 变化路径。
Unaffected by the year selector above — always shows the RE trajectory of selected (or default current-year Top-6) countries across all 24 panel years, 2000–2023.
已选国家用于图表对比
Selected Countries for Chart Comparison
点击上方表格中的"加入对比"按钮，最多选择 6 个国家，下方图表将自动更新。
Click "Add to Compare" in the table above (up to 6 countries); the charts below update automatically.
RE 综合得分对比（柱状图）
RE Composite Score Comparison (Bar Chart)
V / C / F 结构对比（雷达图）
V / C / F Structural Comparison (Radar Chart)
F - C 散点分布（气泡大小 = V，覆盖全部已选国家）
F - C Scatter Distribution (bubble size = V, all selected countries)
从面板数据中观察到的两类轨迹
Two Trajectories Observed in the Panel Data
以下观察均可在上方"时间序列趋势"图中选中对应国家后直接验证，而非脱离数据的叙述性断言。
Both observations below can be directly verified by selecting the relevant country in the "Time-Series Trend" chart above — they are not narrative claims detached from the data.
韩国：制度能力驱动的持续上升
South Korea: A Sustained Rise Driven by Institutional Capacity
在面板数据中，韩国的 RE 综合得分在样本期内呈现较为稳定的上升趋势，这与其治理质量、监管适应性等制度变量的持续改善相吻合，也与其在半导体、电池等战略产业上快速的政策转化能力相一致。这一轨迹为"制度摩擦下降可放大资源转化效率"的理论命题提供了间接的实证支持。
In the panel data, South Korea's composite RE score shows a relatively steady upward trend over the sample period, consistent with sustained improvements in its governance-quality and regulatory-adaptability variables, and with its rapid policy-conversion capacity in strategic industries such as semiconductors and batteries. This trajectory offers indirect empirical support for the theoretical proposition that declining institutional friction amplifies resource-conversion efficiency.
美国：高基准水平下的温和波动
The United States: Mild Fluctuation Around a High Baseline
美国在样本期内的 RE 综合得分长期保持在样本中较高水平，但呈现温和的阶段性波动，这与其政治极化程度上升带来的制度摩擦（F）压力，以及科技与国防产业转化率（C）保持强势之间的张力相吻合。这一"高位波动而非趋势性下降"的模式提示，即使在摩擦上升的环境下，具备强转化能力的经济体仍可维持较高的重配置效率。
The United States maintains a relatively high composite RE score throughout the sample period, but with mild cyclical fluctuation — consistent with the tension between rising institutional friction (F) from political polarization and continued strength in technology/defense conversion rates (C). This pattern of "fluctuation around a high plateau rather than trend decline" suggests that economies with strong conversion capacity can sustain high reallocation efficiency even amid rising friction.
与既有指数的关系、稳健性检验与局限
Relation to Existing Indices, Robustness Checks, and Limitations
与既有治理/竞争力指数的比较
Comparison with Existing Governance/Competitiveness Indices
RE 指数与世界银行治理指标（WGI）、全球竞争力指数等既有测量存在概念上的重叠（均涉及制度质量与政策效能），但 RE 指数的独特之处在于将"反应弹性"（应对冲击的调整速度）与"转化率"（要素投入转化为产出的效率）显式地纳入同一乘法结构，而非仅停留于静态治理质量的单维打分。经检验，RE 综合得分与既有治理质量代理变量呈现出中等强度的正相关，符合理论预期方向，但并非完全重合，表明该指数捕捉到了既有测量未能覆盖的动态维度。
The RE Index conceptually overlaps with existing measures such as the World Bank's Worldwide Governance Indicators and global competitiveness indices (both touch on institutional quality and policy effectiveness), but its distinguishing feature is explicitly embedding "responsive elasticity" (adjustment speed under shocks) and "conversion rate" (input-to-output efficiency) into a single multiplicative structure, rather than resting on a static, single-dimension governance score. Correlation checks show a moderate, theoretically expected positive relationship between the RE composite and existing governance-quality proxies — but not full overlap, indicating the index captures a dynamic dimension not fully covered by prior measures.
稳健性检验
Robustness Checks
构建流程对权重方案（等权 vs. PCA 动态权重）、插补策略（列表删除 vs. 迭代插补）及滚动窗口长度进行了敏感性分析，结果显示各国 RE 排名的相对顺序在不同规格下保持较高的稳定性；自举法估计的置信区间显示，样本量较小或数据缺失较多的国家-年份，其综合得分的不确定性相应更高，提示解读排名差异时应结合置信区间而非仅比较点估计。
Sensitivity analyses varied the weighting scheme (equal weights vs. dynamic PCA weights), the imputation strategy (listwise deletion vs. iterative imputation), and the rolling-window length; relative country rankings remain fairly stable across specifications. Bootstrap confidence intervals show that country-years with smaller effective samples or more missing data carry correspondingly higher uncertainty in the composite score — a reminder to interpret ranking differences alongside confidence intervals rather than point estimates alone.
已知局限性
Known Limitations
- 代理变量效度：V、C、F 均以可观测的宏观/治理变量代理理论构念，代理关系本身依赖理论假设，无法完全排除测量偏误。
- Proxy validity: V, C, and F are each proxied by observable macro/governance variables; the proxy relationship rests on theoretical assumptions and cannot fully rule out measurement error.
- 跨国可比性：不同国家的统计口径、数据采集能力存在差异，插补过程虽可缓解缺失但无法完全消除由此产生的偏差。
- Cross-national comparability: statistical conventions and data-collection capacity differ across countries; imputation mitigates but cannot fully eliminate resulting bias.
- 样本范围：本站发布版本移除了中国数据行，仅覆盖 18 个非中国 G20 成员，任何全球或"G20整体"层面的结论均不适用于本页数据。
- Sample scope: the version published here excludes China's data rows, covering only 18 non-China G20 members; no global or "G20-as-a-whole" conclusion should be drawn from this page's data.
- 时间窗口边界效应：滚动窗口 PCA 权重在样本首尾年份（如 2000–2001、2022–2023）的估计稳定性略低于中间年份。
- Window boundary effects: rolling-window PCA weight estimates are somewhat less stable at the start/end of the sample (e.g., 2000–2001, 2022–2023) than in the middle years.
将真实数据接入计算器
Load Real Data Into the Calculator
如需查看覆盖 OECD 全部 38 个成员国的同类指数、示意性国家案例数据库，或使用手动权重的 RE 计算器探索理论机制，可前往以下页面继续研究。
To explore the parallel index covering all 38 OECD member states, the illustrative country-case database, or use the manual-weight RE calculator to probe the theoretical mechanism, continue to the pages below.