Systems Research系统研究Oct 17, 20262026年10月17日Note讯息

The goods shock is in the pipe, not on shelves.商品价格的冲击已经进了管道，只是还没上货架。

A long steel pipeline runs from a hazy harbor across a frosted plain to a quiet shop with one dim lit window; the joints nearest the shop vent thin steam.一条长长的钢管从雾中的港口穿过结霜的平原，通向一家安静的小店，店里只亮着一扇昏暗的窗；靠近小店的管道接口冒出细细的蒸汽。AI-generated image · concept, not evidenceAI 生成图像 · 概念示意，不作为事实证据

Import prices rose 1.1% in September, and most readers will file that under oil. The fuel line did jump. The number that matters is the one without fuel, and it says the price shock has already left the tank.

Per BLS on Oct 16, nonfuel import prices rose 0.7% in September after a revised 0.6% in August. Over 12 months they are up 6.0%, the largest annual rise since the year ended April 2022.

Nonfuel industrial supplies rose 1.4%, after 1.3%. Capital goods rose 0.9%, a second straight gain. Prices of goods from China rose 0.7%, the most since December 2021.

On the shelf, it is quiet. Core goods in the CPI were up 0.7% from a year earlier in September, per BLS on Oct 14.

We infer the path. An energy shock does not stay in fuel. It moves into materials that take energy to make, then into machines and consumer goods months later. Importers and retailers sit between the dock and the shelf and decide how much to pass on. Until they do, the cost sits in their margins.

The dollar did not do this. The Fed's broad dollar index rose in the week to Oct 9, which should make imports cheaper, not dearer.

Fuel is the headline; materials are the transmission.

The second-order effect: capital goods get dearer right when firms are pricing AI and data-center equipment. The cost of building can rise even if the cost of money falls.

The strongest counter: imports are a small slice of what consumers pay. Distribution, labor and domestic content dilute a 6% dock price into much less at the register, and consumer goods excluding autos rose only 0.4% at the dock. For household prices over the next few months, calm core goods inflation may be the right signal. So this read is for people who buy materials and equipment, not for the grocery bill.

If you buy equipment or materials for a small business, compare current supplier quotes with quotes from three months ago before you read calm CPI as calm costs. One question for this week: what did your last three quotes say?

Wrong if nonfuel industrial supplies and capital goods import prices are both flat or falling in BLS's October report, due Nov 17.

Research, not investment advice.

9 月进口价格上涨 1.1%，多数人会把它归到油价上。燃料价格确实大涨，但关键是去掉燃料之后的那个数字：它说明价格冲击已经离开了油箱。

据美国劳工统计局（BLS）10 月 16 日公布，9 月非燃料进口价格环比上涨 0.7%，8 月修正后为 0.6%；同比上涨 6.0%，是截至 2022 年 4 月那一年以来最大的年度涨幅。

其中，非燃料工业原材料上涨 1.4%（前值 1.3%）；资本品上涨 0.9%，连续第二个月上涨；自中国进口的商品价格上涨 0.7%，是 2021 年 12 月以来最大涨幅。

货架上却很平静。据 BLS 10 月 14 日公布，9 月 CPI 中的核心商品同比只涨 0.7%。

我们推断传导路径是这样的：能源冲击不会停留在燃料上，它先进入需要大量能源才能生产的原材料，几个月后再进入机器和消费品。进口商和零售商夹在码头和货架之间，决定转嫁多少；在转嫁之前，成本压在他们的利润率里。

这不是美元造成的。美联储的广义美元指数在截至 10 月 9 日的一周里是上涨的，这本应让进口变便宜，而不是变贵。

燃料是标题，原材料才是传导。

二阶效应：资本品恰恰在企业为 AI 和数据中心设备定价的时候变贵。即使资金成本下降，建设成本也可能上升。

最强的反方是：进口只占消费者所付价格的一小部分。分销、人工和本地成分会把码头上 6% 的涨幅稀释成收银台上小得多的数字；而在码头环节，不含汽车的消费品只涨了 0.4%。就未来几个月的家庭物价而言，温和的核心商品通胀也许才是对的信号。所以这个判断针对的是采购原材料和设备的人，而不是日常买菜的账单。

如果你为小企业采购设备或原材料，在把“CPI 平静”读成“成本平静”之前，先把现在的供应商报价和三个月前的报价比一比。本周值得回答的一个问题：你最近三次报价是怎么变的？

如果 BLS 11 月 17 日公布的 10 月报告显示，非燃料工业原材料和资本品的进口价格都持平或下跌，这个判断就错了。

研究，不构成投资建议。

Sources来源

WAKENEX / NOTES

Sources来源

READ NEXT接着读

Open weights you can't ship are a trial, not an exit.不能上线的开放权重，是试用，不是出口。

Deleting the expiry date doesn't change the leverage.删掉到期日，杠杆并没有变。

Agents now compete on cheap human steering, not on generation.编码智能体如今比拼的是“让人低成本地掌舵”，因为限制产出的是人的监督注意力，而不是生成能力。