If homo economicus exists anywhere, surely he must be found among financial traders. Amid the shouting and shoving of trading pits and the manic clicking and phone slamming of electronic trading desks, the atomized, rational, self-interested agency of neoclassical economic theory appears to emerge. Ethnographers of traders have refuted atomism and instead proposed a socially embedded but still rational and self-interested agent: an embedded homo economicus. I take this critique a step further and argue that traders are neither rational nor self-interested. Certainly, traders present themselves as rational and self-interested, but a ‘thick’ understanding of their social performances and cultural forms proves otherwise. My analysis focuses on the practice of friendly betting among Chicago traders. Despite its competitive appearances, friendly betting constitutes a non-rational, non-self-interested system of gift exchange and evinces why the myth of embedded homo economicus is both beguiling and false. This conclusion cuts to fundamental issues of agency and practice, and calls for widespread scepticism of theories of rational self-interest. As neoclassical economic theory and market ideology spread around the globe and rational self-interest becomes increasingly normalized, we must be wary of confirmation bias and remember the old ethnographic principle that people aren’t always what they say they are.

如果经济存在于任何地方，肯定必须在金融交易员中找到他。在交易场所的呼喊和推挤，电子交易台的狂躁的点击和电话的轰鸣之中，新古典经济学理论的原子化，理性，自私的代理似乎出现了。交易员的民族志学家驳斥了原子主义，而是提出了一种社会内在但仍然是理性和自私的代理人：一种内在的经济人。我对这一批评进行了进一步探讨，认为交易者既不理性也不自私。当然，交易者表现为理性和自私自利，但是对他们的社会表现和文化形式的“深入”理解则相反。我的分析集中于芝加哥交易员之间的友好博彩实践。尽管外观颇具竞争力，友好博彩构成了一种非理性，非自私的礼物交换系统，这也说明了为什么嵌入的“同质经济”的神话既令人迷惑又是虚假的。该结论涉及代理和实践的基本问题，并呼吁人们对理性的自我利益理论产生广泛的怀疑。随着新古典经济学理论和市场意识形态在全球范围内的传播以及理性的自我利益日益规范化，我们必须警惕确认偏差，并记住古老的人种学原理，即人们并不总是像他们所说的那样。