This article is a supplement to my main contribution to the Routledge Handbook of Complexity Economics (2023). On the basis of three recent papers, it presents an unconventional perspective on economic inequality from a statistical physics point of view. One section demonstrates empirical evidence for the exponential distribution of income in 67 countries around the world. The exponential distribution was not familiar to mainstream economists until it was introduced by physicists by analogy with the Boltzmann-Gibbs distribution of energy and subsequently confirmed in empirical data for many countries. Another section reviews the two-class structure of income distribution in the USA. While the exponential law describes the majority of population (the lower class), the top tail of income distribution (the upper class) is characterized by the Pareto power law, and there is no clearly defined middle class in between. As a result, the whole distribution can be very well fitted by using only three parameters. Historical evolution of these parameters and inequality trends are analyzed from 1983 to 2018. Finally, global inequality in energy consumption and CO2 emissions per capita is studied using the empirical data from 1980 to 2017. Global inequality, as measured by the Gini coefficient G, has been decreasing until around 2010, but then saturated at the level G=0.5. The saturation at this level was theoretically predicted on the basis of the maximal entropy principle, well before the slowdown of the global inequality decrease became visible in the data. This effect is attributed to accelerated mixing of the world economy due to globalization, which brings it to the state of maximal entropy and thus results in global economic stagnation. This observation has profound consequences for social and geopolitical stability and the efforts to deal with the climate change.

翻译：本文是对我主稿《Routledge复杂性经济学手册》(2023年)的补充。基于三篇近期论文，本文从统计物理学角度提出了关于经济不平等的非传统视角。第一部分展示了全球67个国家收入指数分布的实证证据。这种指数分布此前并未被主流经济学家所熟悉，直到物理学家通过类比玻尔兹曼-吉布斯能量分布将其引入，随后在多国经验数据中得到证实。第二部分考察了美国收入分布的两类结构特征：指数律描述了大多数人口（底层阶级），而收入分布的顶端尾部（上层阶级）则遵循帕累托幂律分布，两者之间不存在明确的中间阶级。因此，整个分布仅需三个参数即可完美拟合。本文分析了1983年至2018年间这些参数的历史演变及不平等趋势。最后，利用1980年至2017年的经验数据研究了全球人均能源消耗与二氧化碳排放的不平等性。以基尼系数G衡量的全球不平等在2010年前持续下降，随后在G=0.5水平趋于饱和。这一饱和水平早在全球不平等减缓趋势在数据中显现之前，就已基于最大熵原理在理论上被预测。该效应归因于全球化加速了世界经济的混合，使系统达到最大熵状态，进而导致全球经济停滞。这一观察结果对社会与地缘政治稳定以及应对气候变化的行动具有深远影响。