Posen Sees Fed Raising Rates 100 Basis Points on Sticky Inflation

波森预计通胀粘性将促使美联储加息100个基点 Adam Posen表示，由于持续的核心服务通胀将把物价涨幅压制在3.5%以上，美联储在未来六个月内可能加息75至100个基点。 Adam Posen对Kevin Warsh主席的杰克逊霍尔演讲进行了评估，并得出结论：持续的核心通胀很可能会迫使美联储在未来数月内加息75至100个基点。讨论批评了Warsh推崇个人自由裁量权而非基于规则框架的做法，审视了FOMC内部的制度规范，并重点指出了财政主导等对央行独立性日益上升的结构性威胁。此外，Posen还深入分析了经济学家流向顶尖AI实验室的趋势，并解释了“混乱工作”（messy jobs）理论以及消费者对人际互动的偏好如何推迟了AI驱动的大规模劳动力替代。 Adam Posen对Kevin Warsh的杰克逊霍尔演讲给出了扎实的“B”级评价，指出尽管讲话避免了明确指引，但其论点的逻辑走向暗示了即将收紧货币政策的必要性。 (3:45) Warsh强调诸如通胀以“合适的速度朝正确的方向”移动等主观标准，这是以牺牲制度化、基于规则的框架为代价，将个人自由裁量权最大化。 (6:35) 依赖个人判断而非系统性的通胀目标制，让人联想起Alan Greenspan在20世纪90年代后期的任期，并可能引发“老板之死”（Weekend at Bernie's）式的脆弱性，即市场稳定性过度依赖于单一个人。 (9:55) 通胀连续64至65个月高于目标水平，源自美联储具体的政策失误——包括推迟加息和过早降息——并被缺乏资金支持的大规模财政刺激措施进一步放大。 (14:15) 当财政主导迫使压低利率以为政府赤字融资，或者政治力量试图重构美联储治理架构和地方联储投票权时，央行独立性将面临最大威胁。 (20:00)

Adam Posen, President of the Peterson Institute for International Economics, warns that the Federal Reserve faces stubborn inflation driven by sustained fiscal indiscipline, predicting that the central bank will need to raise interest rates despite rising political friction and institutional constraints. Decoding Warsh's Jackson Hole Hawkish Shift (3:45) Assessing Federal Reserve Chair Kevin Warsh's headline Jackson Hole address, Posen awards the speech a "B" grade (potentially a "B-plus" if it successfully marks an inflection point past early communication missteps), noting that attendees interpreted its comprehensive review of persistent inflation pressures as an implicit signal to tighten policy. Warsh explicitly reaffirmed a 2% core PCE inflation target and pushed back against using wage growth as a leading inflation indicator. However, Posen criticizes Warsh’s pursuit of personal discretion over systematic rules. By stating policy depends on inflation moving in the "right direction and at the right speed," Warsh adopted an ambiguous standard reminiscent of Alan Greenspan’s late-1990s personal-judgment model, diverging from the rules-based monetary frameworks championed by his mentor John Taylor. Fed Independence and Fiscal Dominance Risks (20:00) Posen emphasizes that central bank independence exists to protect the long-term credibility of U.S. government debt, rather than to suppress borrowing costs to accommodate short-term Treasury bond issuance. While policymakers have historically weathered public complaints from presidents, Posen identifies severe structural threats, including proposals to strip regional Federal Reserve Bank presidents of their votes, shorten their tenures, rewrite the dual mandate, or curb the Fed's budget. Internally, Posen argues that former Chair Jerome Powell's drive for consensus during the pandemic and political attacks created institutional inertia, with 12 to 15 consecutive meetings without a dissent. That elevated barrier to open debate left the committee slow to respond to the inflationary consequences of unfinanced tax cuts and aggressive fiscal stimulus under both the Trump and Biden administrations. Sticky Inflation and Projected Fed Rate Hikes (60:00) Posen forecasts that the Federal Reserve will raise the federal funds rate by 75 to 100 basis points over the next six months. He anticipates an interest rate hike in either September or December, viewing a December hike as nearly certain even if the Fed moves in September. While headline figures may occasionally drop on volatile energy prices, core services inflation excluding imports is accelerating into the high 3% range across three-, six-, and twelve-month moving averages, leaving inflation likely stuck between 3.5% and 4.5% in the near term. Posen attributes the ongoing stretch of 64 to 65 months of above-target inflation to policy errors under Powell—notably delaying rate hikes in 2022, cutting rates too early, and relying on incorrect forecasts of labor market collapse—compounded by the lagged supply shocks of tariffs and restrictive immigration policies. The AI Labor Paradox and "Messy Jobs" (40:00) Examining artificial intelligence, Posen points out that despite elite macroeconomists migrating to frontier labs like Anthropic and OpenAI, official payroll data shows early productivity indicators rather than aggregate labor displacement. Hiring has continued even in occupations widely expected to face immediate cuts, such as entry-level coders and truck drivers. Posen attributes this dynamic to Erik Brynjolfsson's productivity "J-curve"—a transitional multi-year lag during which organizations restructure workflows around hybrid human-AI teams—and Luis Garicano's "messy jobs" concept, which underscores that real-world employment relies on tacit knowledge, physical nuance, and relational context. Consequently, Posen argues that AI-driven macroeconomic productivity gains have only begun registering over the past 18 months, constrained by consumer preferences for human interaction. Posen concludes that if the Federal Open Market Committee fails to deliver interest rate hikes after signaling persistent inflation, markets may interpret the hesitation as backing down under White House pressure. Meanwhile, parallel efforts—such as the communications task force led by Mervyn King and Peter Fisher—suggest central banks may soon be forced to scale back explicit forward guidance to restore independent market price discovery.

彼得森国际经济研究所所长亚当·波森指出，新任美联储主席凯文·沃什在杰克逊霍尔的讲话隐性传递了紧缩意图，但其偏好个人自由裁量权的治理取向与持续顽固的通胀正将美联储带入险境。波森预计，在核心服务通胀粘性与财政刺激的共振下，美联储未来六个月内或需累计加息75至100个基点，而央行独立性与宏观经济研究正面临政治干预及科技转型的多重挑战。 沃什首秀：紧缩信号与自由裁量权争议 (3:45) 在杰克逊霍尔年会上，新任美联储主席凯文·沃什发表了备受关注的演说。波森对其常态评分为“B”（若算上初期沟通失误后的修正则为“B+”）。尽管沃什重申了2%的核心PCE通胀目标并质疑将薪资作为通胀领先指标的合理性，但欧洲央行官员及与会者普遍将其讲话解读为紧缩信号。沃什罗列了通胀可能反弹的多重依据，逻辑结论即为货币政策“必须收紧”。波森警告，若FOMC后续未能加息，市场将质疑沃什是否在白宫政治压力下临阵退缩。 更深层的担忧在于沃什追求最大限度的个人裁量空间。沃什主张通胀须朝“正确方向和速度”靠拢，却未提供明确量化时间表。波森认为这种模糊立场使政策重回逐次会议相机抉择的状态，形同1999年格林斯潘式的个人声望治理，面临严重的“伯尼周末”脆弱性，背离了伯南克等人推动的基于规则与制度化通胀目标的问责体系。 委员会共识惯性与央行独立性面临的威胁 (20:00) 针对美国连续64至65个月通胀超标，波森认为鲍威尔领导下的美联储难辞其咎，其失误包括2022年加息过迟、前期降息过早以及低估劳动力市场韧性。此外，拜登政府的财政刺激与特朗普早期的未对冲减税共同强化了通胀势头。然而，在FOMC内部，鲍威尔在应对疫情与政治攻击期间筑起了极高的公开异议门槛，导致委员会连续12至15次会议毫无反对票，形成了抑制辩论的制度惯性。 波森强调，央行独立性的核心是捍卫美国国债的长期发债信誉，而非压低当届政府的发债成本。特朗普时期的外部干预已演变为体制威胁，包括提议剥夺地区联储主席投票权、推动政治任命、试图通过国会拨款限制美联储预算等。波森警告，一旦财政部或白宫迫使央行人为压低利率以消化国债，将触发最具宏观破坏力的“财政主导”危机。 央行沟通机制反思与前瞻性指引的局限 (28:30) 沃什设立了由前英格兰银行行长默文·金与前纽约联储官员彼得·费舍尔牵头的沟通工作组。波森预计该小组将提出激进建议。1992年英国退出欧洲汇率机制后，金曾首创通胀目标制与概率分布扇形图，树立了透明度标杆；但近年来，金与费舍尔转而对前瞻性指引持严厉批评态度，认为央行预测能力有限，频繁发布嘈杂预测反而误导市场。 该批评指出，向市场过度投喂政策路径削弱了投资者的自主风险定价能力，被人为压低的收益率波动容易催生资产泡沫和道德风险。沃什在讲话中亦承认，过度顺应短期市场噪音会使央行陷入“镜像大厅”式的回音室效应。工作组的改革方向预示着央行或将减少前瞻指引，推动债券市场恢复双向风险发现机制。 “棘手工作”假说与AI宏观生产率现实 (40:00) 针对生成式AI引发的技术性失业恐慌，波森援引PIIE及相关学者的实证研究指出，官方数据尚未显示出预期的失业潮，初级程序员与长途卡车司机的就业人数仍保持增长。布林约尔松提出的“J曲线”理论表明，技术扩散需经历约五年的业务重组与人机协作磨合期；同时，加里卡诺的“棘手工作”假说揭示出，现实劳动力包含大量算法难以替代的隐性知识、线下互动与现场判断。 此外，公众对真人互动的消费偏好为技术替代设立了上限。波森认为，AI对全要素生产率的实质拉动在过去18个月才初见端倪。尽管英伟达等芯片设计外包导致美国GDP统计可能存在约0.3%的漏损，且未计入消费者剩余，但宏观红利的兑现仍取决于传统产业对新技术的流程重塑。波森还对知名经济学家纷纷加盟AI企业表达忧虑，认为高额薪酬与企业从属削弱了学术研究的独立性。 核心通胀粘性与美联储潜在的加息路径 (56:56) 回顾此前与彼得·奥斯泽格预判通胀触及4%的研究，波森指出鲍威尔团队的盲区在于误判了劳动力市场韧性、金融环境宽松度以及财政刺激与关税移民政策的滞后影响。由于通胀已高企五十多个月，涨价动能深植于经济体系中，极易受供给侧与地缘动荡冲击。 波森指出，当前排除进口商品的核心服务业通胀在3个月、6个月和12个月均线维度上均向高3%区间攀升，显示出高度粘性。波森预测美联储将被迫采取行动：若9月不加息，12月加息几成定局；即便9月加息，12月仍大概率跟进。未来六个月内联邦基金利率或需提高75至100个基点，在此之前，通胀率恐持续在3.5%至4.5%区间徘徊。 鲍威尔任期留下了长期通胀超标的治理遗产，而新任主席沃什的鹰派表态虽指向紧缩，却暗含倒退回个人高度裁量权的不确定性。面对日益临近的财政主导风险、顽固的核心服务通胀以及尚未实质兑现的AI生产率红利，美联储若继续以等待数据为由延宕紧缩决断，将可能被迫在未来实施更剧烈的补救性加息，并进一步损害央行的制度信誉。

## Background 本期节目由彭博社（Bloomberg）旗下的知名播客《Odd Lots》于2026年9月1日在RSS及各大播客平台发布。节目简介显示，主持人在杰克逊霍尔（Jackson Hole）央行年会现场采访了彼得森国际经济研究所（PIIE）所长亚当·波森（Adam Posen），围绕美联储新任主席凯文·沃什（Kevin Warsh）的偏鹰派演讲、美联储未来的加息路径、财政主导与央行独立性受到的政治干预、AI对就业与“混乱工作”的真实冲击等核心议题展开深入探讨。 ## TL;DR 波森给沃什在杰克逊霍尔的演讲打出了“B”的评分，指出演讲逻辑虽隐晦指向必须加息，但沃什执意追求“最大化个人自由裁量权”而拒绝明确指引，存在巨大的政策风险与治理倒退隐患。波森批评了鲍威尔任期内导致通胀长达65个月超标的一系列迟缓与误判，并预测核心通胀粘性将迫使美联储在未来半年内累计加息75至100个基点。此外，嘉宾与主持人们还深入剖析了默文·金主导的沟通工作组可能带来的激进变革、经济学界投奔AI巨头引发的独立性隐忧，以及现实中“混乱工作”（Messy Jobs）对AI替代所形成的天然壁垒。 ## Atmosphere & style 现场对话直率犀利且充满专业洞见，波森一改央行圈常见的委婉辞令，以业内“局内人”的笃定与辛辣风格实话实说、不留情面。两位主持人与波森的互动默契风趣，既有对格林斯潘、伯南克时代掌故与美联储决策潜规则的深度解密，也有结合翻修屋顶、派对钢琴师等生活趣事的幽默拆解。整场访谈在学术严谨的宏观推演与生动逼真的现实讽喻之间自如切换，听感兼具高度的智性交锋与畅快淋漓的揭秘感。 ## Bulletin - `03:45` 亚当·波森为美联储新主席沃什在杰克逊霍尔年会的首秀演讲打出“B”级评分，指出演讲逻辑虽强烈暗示需要加息，但沃什极力追求“相机抉择”的模糊表态也增加了美联储落后于通胀曲线的风险。 - `01:00:00` 针对当前通胀走势，波森预测核心服务业通胀极具粘性，预计通胀率将在 3.5% 至 4.5% 区间徘徊，美联储在未来 6 个月内不得不累计加息 75 至 100 个基点。 - `27:00` 波森提醒市场关注由默文·金主导的美联储沟通特别工作组，认为其对现行前瞻性指引和预测模型的根本性怀疑，可能比拉詹牵头的资产负债表小组带来更具颠覆性的机制改革。 - `33:30` 针对顶尖经济学家大规模流向 AI 科技巨头的现象，波森警告学界的独立客观性正在被高薪与产业叙事侵蚀；同时“混乱工作”假说表明，因人机协同与复杂情境知识，短期内 AI 尚未引发广泛失业。 ## Key takeaways - `05:00` 沃什在杰克逊霍尔的演讲虽然声称反对前瞻性指引，但其实质逻辑已在事实上给出了必须紧缩收紧政策的强烈暗示。 - `10:00` 沃什试图保留“最大化自由裁量权”且拒绝承诺具体门槛指标的做法，让学界担忧美联储可能重回格林斯潘时代缺乏制度化动责的个人化决策模式。 - `12:08` 美联储内部拥有“绝不公开否定主席”的潜规则与强共识惯性，沃什口中提倡的“善意家庭争吵”在实际操作中面临巨大的制度文化阻力。 - `14:48` 波森认同鲍威尔治下经历长达 65 个月通胀超标是一次严重的政策失误，其根源在于 2022 年加息过迟、过早启动降息以及未能及时对冲拜登时期的顺周期财政扩张。 - `18:28` 央行独立性的真正生死防线在于抵御“财政主导”：当白宫与财政部要求央行压低收益率配合发债时，合格的央行官员必须坚决说“不”。 - `27:00` 相比市场高度瞩目的资产负债表小组，默文·金与彼得·费舍尔领导的沟通小组更具颠覆潜力，他们倾向于打破央行过多预测噪音并容忍更高的市场波动率。 - `33:30` 大量顶尖经济学家加入头部 AI 巨头参与公共政策设计，使得学术研究与独立声音面临潜在的信誉危机和自我审查风险。 - `40:00` 实证数据至今未见 AI 导致岗位替代潮，布林约尔松的“J 曲线效应”与加里卡诺的“混乱工作”理论解释了人机协同及深层隐性常识难以被算法瞬间取代的现实。 - `50:00` 英伟达等科技巨头海外代工导致的约 0.3 个百分点 GDP 漏算争议虽不改长期宏观大势，但折射出传统国民经济核算在衡量数字革命与生产率跃迁时的结构性困境。 - `55:00` 个人微观效用的提升并不直接等同于宏观生产率爆发，AI 带来的真正生产率红利必须经历传统行业彻底重构业务流后才会显现。 - `01:00:00` 非进口商品的核心服务业通胀粘性极强，由于前期财政刺激滞后发酵及金融条件尚未真正收紧，美联储预计在未来半年内需要加息 75 至 100 个基点。 - `01:01:45` 公众与历史对高通胀的厌恶远甚于失业率，鲍威尔任内长期的通胀失控使其历史声誉面临被严厉重估的风险。 ## Detailed walkthrough ### `00:00` 杰克逊霍尔年会的开场与嘉宾介绍 节目一开始，主持人 Joe Weisenthal 与 Tracy Allaway 介绍了本期《Odd Lots》的背景。他们再次来到怀俄明州杰克逊霍尔（Jackson Hole）参加堪萨斯城联储举办的年度央行年会。讨论首先从央行官员特有的交流风格切入：Joe 感慨在央行圈子里，官方场合的交流往往极度礼貌与克制，大家说话总是充满“外交辞令”（diplomatic），以至于很难听出他们内心深处究竟在想什么、对经济形势的真实感受如何。因此，在杰克逊霍尔这样的场合，能找到一位不拐弯抹角、愿意直言不讳（give it to us straight）的专家显得尤为珍贵。 随后他们隆重迎来了彼得森国际经济研究所（PIIE）所长、前英国央行货币政策委员会（MPC）委员亚当·波森（Adam Posen）。Joe 打趣道，波森正是那种绝不惧怕在公众面前大谈“财政主导”（fiscal dominance）等敏感话题、总是实话实说的人。由于这是美联储新任主席凯文·沃什（Kevin Warsh）在本次杰克逊霍尔发表演讲后录制的首期节目，话题自然直奔主题：波森究竟如何看待并评价沃什主席的这篇重磅演讲？ --- ### `03:45` 给沃什演讲打分与初期的“成长阵痛” 面对打分要求，波森给出了一个“B”的评分。他解释说，如果按常规时期的正常标准来看，这仅仅是一份中规中矩的 B 级演说；但结合美联储近几个月所处的动荡背景，这份演讲的实际效果要积极得多。在此之前，这位新任主席不仅在央行同行之间，也在专业市场观察家当中引发了巨大的困惑——大家完全摸不清他内心究竟想做什么。无论在短期内是否打算加息，还是中长期内设立的诸多工作组（task forces）与宏观表态背后究竟有何意图，局势都扑朔迷离；更糟糕的是，沃什在 7 月份的新闻发布会上彻底搞砸了沟通。 不过波森客观指出，在美联储历任主席中，早期发布会栽跟头并不罕见。格林斯潘（Greenspan）、耶伦（Yellen）和鲍威尔（Powell）这四位前任中的三位都在履职初期有过重大失误。波森认为，如果六个月后回看，能确认本次演讲是一个真正的政策拐点——即沃什前几个月的混乱表态仅仅是新官上任的“成长阵痛”（growing pains），随后他被残酷的现实狠狠敲醒并重回正轨——那么这篇演讲完全可以上调至“B+”。此外，波森观察到本次年会的气氛与去年大不相同：去年全场笼罩着对美联储独立性受到攻击的焦虑，鲍威尔作为主席参加的最后一届杰克逊霍尔甚至赢得了全场起立鼓掌；而今年，沃什及其团队显然有意将一切拉回“常规化”，从现场反响来看，这次年会确实恢复了平静与常态。 --- ### `05:00` 鹰派前瞻、核心锚定与走钢丝的“最大化自由裁量权” 谈及演讲的实质内容，波森援引了一位欧洲央行官员在早餐时的评价：尽管沃什嘴上极力宣称自己反对前瞻性指引（forward guidance），但这篇演讲实际上就是在告诉市场——美联储必须加息。整篇演讲分为四个部分，第四部分巨细靡遗地罗列了通胀为何难以稳定、甚至可能长期居高并进一步反弹的各种原因。虽然沃什没有直接说出最后那句结论，但逻辑链条的终点明摆着就是“因此我们不得不收紧政策”。波森警告称，这种强烈的加息暗示也带来了极大的政策风险：如果美联储接下来没有真正加息，外界就会质疑是不是因为来自特朗普（Trump）的政治压力，或者主席在委员会中“冲得太快而脱节”（out over his skis）。 ```mermaid flowchart TD accTitle: 沃什演讲逻辑链条与市场隐忧 accDescr: 展示沃什演讲中通胀论据到政策暗示，以及潜在失信风险的逻辑流程 subgraph Analysis["演讲第四部分逻辑推演"] A["列举通胀顽固与上行风险"] --> B["逻辑推论：政策必须紧缩收紧"] B --> C["实际未明说：'因此我们必须加息'"] end subgraph Risks["若后续未能兑现加息"] C -.-> D["市场质疑是否受到政治施压 (Trump因素)"] C -.-> E["主席在 FOMC 委员会中冒进脱节"] end ``` 在细节澄清上，沃什也纠正了此前数月里的一些混乱表述，波森对此予以肯定： - 重新明确将核心 PCE（Core PCE）达到 2% 作为法定通胀目标，消除了此前对于“通胀定义模糊”的摇摆。 - 明确指出工资通胀并不是预测整体通胀的良好领先指标，波森认为这是一个完全站得住脚的常规央行观点。 - 承认金融状况（financial conditions）在政策制定中的重要性。 尽管有这些积极修补，波森依然指出了演讲的两大硬伤。首先，演讲第二部分长篇大论阐述其“个人原则”，内容空洞甚至充满陈词滥调，不仅没能答疑解惑，反而让人困惑于他的真正意图，完全违背了反对前瞻性指引应有的简练风格。其次也是最令学界与前央行官员深感忧虑的，是沃什在演讲中极力试图保留“最后的自由裁量权”（maximize his last discretion）。例如在演讲结尾，沃什提出通胀必须朝着“正确的方向”以“正确的速度”回落，但从未明确何为“正确的速度”。波森与 Tracy 一致认为，没有速度约束的方向毫无意义，这实质上让美联储退回到了过去四年的模糊状态；从听证会、首批新闻发布会、欧洲央行辛特拉（Sintra）会议到本次演讲，沃什最一贯的做法就是拒绝事前承诺任何具体指标，执意保留在每次议息会议前最后一刻随性做决定的特权。 ### `10:00` 规则与相机抉择：从格林斯潘的“网球场危机”到通胀目标制 波森（Adam Posen）指出，沃什（Kevin Warsh）如今的政策倾向显得颇为反常——他出自胡佛研究所（Hoover Institution），一向被视为保守派，而且师从已故的著名经济学家约翰·泰勒（John Taylor，泰勒规则的提出者）。在中央银行学的经典坐标系中，决策者通常在“规则”（rules）与“相机抉择”（discretion）之间寻找平衡点。规则的极端就像只要货币供应量上升就机械地加息，如果执行过于僵化往往带来糟糕后果；而另一个极端则是纯粹的相机抉择，即决策者自视不需要证明决定的逻辑，只凭直觉就能在每次都做出正确判断。 这恰恰是1999年前后格林斯潘（Alan Greenspan）时期的写照。格林斯潘当时之所以能推行极端的个人相机抉择，一方面是因为他个人能力极强，另一方面是因为他对整个联邦公开市场委员会（FOMC）拥有绝对的主导地位。但这为日后埋下了巨大隐患，导致了所谓的“周末狂欢”（Weekend at Bernie's）困境——如果在1999年，年事已高的格林斯潘不幸在美联储网球场上突发心脏病去世，美联储的所有信誉都会瞬间化为乌有，因为外界的信任完全绑架在他个人身上。 正因如此，本·伯南克（Ben Bernanke）、里克·米什金（Rick Mishkin）、托马斯·劳巴赫（Thomas Laubach）以及波森本人才极力倡导“通胀目标制”（inflation targeting）。伯南克出任美联储主席后，非常清醒且刻意地推行动责制与制度化透明度，目的就是不让央行声誉维系于一人之身。波森认为，外界对此讨论严重不足，而这恰恰是沃什目前政策构想中最令人担忧的转向。 ```mermaid flowchart TD accTitle: 货币政策决策机制对比 accDescr: 对比规则主义、相机抉择与通胀目标制三种美联储治理范式 subgraph R1["泰勒规则模式 (Rules)"] A["机械锚定指标<br/>(如货币增长/产出缺口)"] --> B["预设加息/降息路径"] end subgraph R2["格林斯潘模式 (Discretion)"] C["高度依赖个人权威与直觉"] --> D["'相信我能做对决策'"] D --> E["隐患: 个人离场导致信誉崩塌"] end subgraph R3["伯南克/通胀目标制 (Accountability)"] F["制度化通胀目标锚"] --> G["集体动责与透明化沟通"] end ``` ### `12:08` 美联储主席的角色与委员会文化：“家庭争吵”与绝对权威 当被问及美联储主席应当是达成共识的协调者还是强势推行意图的领导者时，波森结合自己在英格兰银行（Bank of England）的任职经历指出，美联储的体系从未明文规定主席的具体职权定位，其风格随不同时期的人选、政治环境以及宏观经济形势而剧烈变动。 不同央行的治理文化截然不同： - **欧洲央行（ECB）**：尽管行长拉加德（Christine Lagarde）备受瞩目，但欧洲央行本质上是一个高度追求共识的机构，决策必须尽可能让各成员满意，这很大程度上是因为委员们代表着不同的主权国家，而非单一国家内部的各个行政区。 - **英格兰银行（BOE）**：文化核心在于开放的激烈探讨——也就是沃什近期反复提及的“善意辩论”或“家庭争吵”（family fight），英格兰银行极其推崇这种充分交锋的决策机制。 - **美联储（Fed）**：至少自沃尔克（Paul Volcker）以来的过去45年里，美联储传统上赋予主席极高的权力与顺从度。虽然伯南克曾试图推动更具讨论氛围的开放式会议，但2008年金融危机打断了这一进程。当时受制于美国苛刻的信息公开法规（即委员会成员超过4人同场开会就必须向公众披露并制作详尽记录），危急关头没有时间走程序，最终决策权实际上完全集中在由伯南克主席、纽约联储主席、沃什以及副主席唐·科恩（Don Kohn）组成的四人核心圈子里。 美联储委员会的固有伦理和潜规则是“绝不能公开否定主席”。正如前美联储理事拉里·迈耶（Larry Meyer）曾指出的，委员会内部存在一条非成文惯例：在任何单次会议上，提出异议（dissent）的人数都有严格上限，而且主席永远不能在投票中落败。当年即便是沃尔克，一旦发现自己有在投票中被否决的苗头，也会直接放下狠话：“下一次会议就是我的最后一次会议。”这也正是波森对沃什强调的纯粹“相机抉择”感到忧虑的原因——在美联储的文化惯性下，这种模式往往会演变成主席个人意志的独断。 ### `14:48` 评判鲍威尔任期：65个月通胀超标与美联储的判断失误 针对沃什关于“美国面临长达64或65个月的通胀超标，且这完全是美联储过错”的批评，波森认为这一论断基本成立。尽管特朗普在提名沃什期间的舆论造势中充斥着火药味，沃什早期关于“要清理门户、整顿美联储”的说辞也显得刻薄且毫无必要，但就事实层面而言，对鲍威尔（Jerome Powell）领导层在委员会顺从下的政策失误提出批评是完全客观合理的。 鲍威尔时期的政策节奏存在三大核心失误：2022年加息启动严重滞后；去年的多次降息即便不能被称为极度荒唐，也绝对属于为时过早；如今则再次落后于曲线（behind the curve）。如果控制进口能源依赖度、危机前的基础通胀率以及能源重要性这三个关键变量做严谨的跨国比较，美联储的表现并不比其他主要央行更光彩。 更为关键的是，美国的通胀压力在很大程度上源于拜登政府上台后推出的无节制财政扩张，且后续毫无财政整顿动作；加之特朗普第一任期推行的长期减税政策，均加剧了需求过热。波森强调，如果认同沃什所说的“通胀归根结底取决于美联储”，那么这种缺乏纪律的财政赤字扩张反而应该构成美联储更加坚决加息的充分理由。 美联储在预测上也出现了系统性偏差。在此前的杰克逊霍尔会议（Jackson Hole）上，美联储工作人员与官员们不断向外界传递“劳动力市场即将大幅走弱、通胀必将自然回落”的观点。当时包括波森本人、美国企业研究所（AEI）的迈克尔·斯特兰（Michael Strain）以及毕马威（KPMG）的黛安·斯旺克（Diane Swonk）等人在内的一批学者就明确警告该判断有误，指出劳动力市场绝不会在未来两年内崩盘，但美联储依然固执己见，最终事实证明鲍威尔团队确实彻底看错了经济走向。 ### `18:28` 政治干预的边界：从常规“唱双簧”到财政主导的真正危险 谈到政治干预美联储的界限，波森区分了“常规政治杂音”与“毁灭性干预”的本质不同。在成熟的民主政治中，白宫在公开场合抱怨美联储或把利率问题当作政治替罪羊本就是博弈的一部分，在某种意义上甚至是一种健康的默契。 在沃尔克、格林斯潘、伯南克乃至耶伦（Janet Yellen）时期，政治高层往往心照不宣：总统面对通胀不得不承受加息代价，为了免于选民谴责，会在镜头前高调痛骂美联储推高利率，但私下里却在“眨眼示意”，默许央行坚决履行抗通胀职责。然而，这种常规政治姿态与特朗普的行为有着天壤之别——特朗普在推特上公然攻击美联储主席是“国家公敌”（enemy of the state），甚至动用法律武器（lawfare）威胁打击理事丽莎·库克（Lisa Cook）甚至鲍威尔本人，这种行径不仅逾越了底线，甚至可以说是违法的。 波森指出，真正最具杀伤力、触及央行独立性底线的危机，在于“财政主导”（fiscal dominance）。当财政部长或总统正竭力推销庞大的美国国债，并直接向美联储施压，要求其“无论通胀有多高，都必须协助压低收益率卖出国债”时，真正的考验便到来了。在此关头，任何一名合格的央行官员、美联储主席或理事，其唯一正确的专业职责就是坚决顶住压力，明确说“不”。 ### `20:00` 维护国债信誉与抵御美联储政治化干预 亚当·波森（Adam Posen）强调，央行官员对财政部的真正职责，是协助财政部及其继任者在长期内可持续地发售美国国债。这意味着央行绝不能通过眼前的操纵、通胀化手段或政策“作弊”来配合短期财政需求；捍卫美国债务的长期信誉，必然要求央行行长有底气顶住行政与财政部门的施压。波森指出，当前经济层面最显而易见的威胁是央行在财政压力面前妥协（即财政赤字货币化或财政主导），但另一项在前总统特朗普执政时期变本加厉、性质同样恶劣的威胁，是对美联储制度结构的公然侵蚀与政治化攻击。 除了针对个别美联储官员的人身指控与违规解雇威胁外，行政部门还曾大肆鼓吹直接削夺部分地方联储行长的投票权，或迅速将他们撤换为政治任命官员，以此打破旨在保障独立性的交错任期制。此外，白宫和国会还一度讨论强行修改美联储的政策目标（法定职责）以及削减、控制美联储的运营预算。波森认为，通过破坏任命机制来剥夺美联储的职能独立性，正是逼迫央行最终屈服于财政压力的政治手段。 - **长期国债信誉的核心**：央行的天职是确保未来几十年美国国债依然具有全球公信力，而不是为了迎合当届财政部长的短期融资便利而扭曲货币政策。 - **美联储体制遭遇的结构性冲击**：削弱地方联储主席投票权、缩短或废除交错任期、替换为政治效忠者、削减美联储预算、强改通胀与就业双重使命。 --- ### `24:30` 鲍威尔时期的共识文化与沃什所称的“家庭争吵” 谈话转向联邦公开市场委员会（FOMC）内部的决策文化。主持人乔·威森塔尔（Joe Weisenthal）提到，他在上一届杰克逊霍尔年会上曾询问芝加哥联储主席奥斯坦·古尔斯比（Austan Goolsbee）为何美联储内部极少出现异议投票（dissents），古尔斯比当时将此归功于杰罗姆·鲍威尔（Jerome Powell）极其出色的共识凝聚能力。这引发了一个尖锐疑问：鲍威尔时期的美联储是否由于“过于擅长”达成一致，以至于FOMC会议室失去了作为真正政策辩论场所的功能？当新任主席凯文·沃什（Kevin Warsh）宣称希望FOMC内部展开一场“健康的家庭争吵”（a good family fight）时，他是否真的比前任更能容忍公开分歧？ 波森对此直言不讳：“我认为沃什是在向外界传递‘我接受更多异议与辩论’的信号，但坦白说，我不相信他。”不过，抛开对沃什个人动机的怀疑，波森承认沃什在道理上是完全正确的——美联储近年的内部辩论确实变得过于稀少，而适度的公开辩论对央行至关重要。波森分析，鲍威尔抑制异议并非出于恶意，而是具备正当的现实背景：先是遭遇前所未有的新冠疫情恐慌，随后又面临来自特朗普政府的猛烈政治围攻。在内忧外患之下，展现公开分裂不仅会引发金融市场恐慌，更会给那些试图攻击美联储独立性的政客送上口实。因此，鲍威尔极大地拔高了提出异议的门槛，倡导全员团结一致。 波森以自己在英国央行货币政策委员会（MPC，2009年至2012年）亲历全球金融危机时的经历作为佐证。当时围绕量化宽松（QE）的争议巨大，时任央行行长默文·金（Mervyn King）在投票前对委员会坦言：“QE确实争议极大，大家可以按自己的意愿自由投票；但一旦表决通过开始执行，我们能否达成暂时的休战协议，不要在公开场合不断质疑QE到底有没有用？”波森当时完全支持这一决定，因为危机当头必须维持政策公信力，而不是让官员在去格拉斯哥演讲时随性扯开话题，沉溺于“按维克塞尔（Knut Wicksell）的货币理论来看如何如何”的学院式空谈。然而，这种危机时期的团结机制一旦形成惯性，就会产生负面效应：当委员会连续12次乃至15次会议全票通过后，提出异议的心理门槛会被无形抬高，最终导致美联储进入一个异议过于罕见的僵化期。 ```mermaid flowchart TD accTitle: FOMC 内部共识惯性的形成机制 accDescr: 展示危机环境与外部政治压力如何通过推高异议门槛，最终形成长期缺乏政策辩论的共识惯性 subgraph 外部冲击与危机应对 A["外部危机爆发<br/>(COVID疫情恐慌 / 政治攻击)"] --> B["领导层诉求全员团结<br/>(避免市场恐慌，抵御政治干预)"] end subgraph 决策文化演变 B --> C["异议投票心理门槛大幅推高"] C --> D["连续十余次会议全票通过"] D --> E["形成共识路径依赖<br/>(内部政策辩论空间萎缩)"] end subgraph 沃什的新诉求 E -.-> F["沃什主张打破沉默<br/>倡导内部‘家庭争吵’式辩论"] end ``` --- ### `27:00` 央行沟通特别工作组：默文·金与可能颠覆预期的改革 针对沃什委派前英国央行行长默文·金（Mervyn King）牵头新成立的美联储“央行沟通特别工作组”，主持人询问这一举动究竟意在解决什么根本问题。波森透露，该三人工作组中有两人与他私交甚笃：默文·金曾是他供职英国央行时的顶头上司；另一位成员、曾任纽约联储高管与美国财政部高级官员的彼得·费舍尔（Peter Fisher），则是彼得森国际经济研究所（PIIE）的现任董事会成员。 波森给出了一个出乎市场意料的判断：外界目前将几乎所有注意力都集中在拉古拉姆·拉詹（Raghuram Rajan）、杰里米·斯坦（Jeremy Stein）和卡伦·戴南（Karen Dynan）组成的“美联储资产负债表特别工作组”上，预期他们会带来颠覆性的资产负债表调整方案；但根据这批经济学家的务实风格，资产负债表小组最终出台的建议很可能是温和、稳妥且小规模的。相反，由默文·金领导的“沟通特别工作组”才最有可能爆出冷门、提出激进的变革方案。 波森解释称，从公开言论便可看出，默文·金和彼得·费舍尔近年来对主流央行现行的沟通策略与前瞻性指引（Forward Guidance）抱有极深的怀疑态度。具有讽刺意味的是，默文·金本人正是通胀目标制（Inflation Targeting）的开山鼻祖之一。早在1992年英国遭遇黑色星期三、英镑被迫退出欧洲汇率机制（ERM）引发严重信任危机时，时任英国央行首席经济学家的默文·金迅速组织骨干团队，借鉴少数先行小型经济体的经验，在英国成功落地通胀目标制，并亲手创立了展示预测不确定性区间的著名工具——扇形图（fan charts）。如今，由这位现代央行沟通范式的奠基人重新审视沟通机制，极有可能打破数十年来由美联储精心构筑的信息指引模式。 - **资产负债表工作组**：成员包括拉詹（Raghuram Rajan）、杰里米·斯坦（Jeremy Stein）、卡伦·戴南（Karen Dynan）；市场预期高，但波森预计其政策建议将偏向克制与实用主义。 - **沟通特别工作组**：由默文·金（Mervyn King）与彼得·费舍尔（Peter Fisher）主导；两人对泛滥的央行沟通和过度前瞻指引持激进怀疑态度，变革力度可能大幅超出市场预期。 - **默文·金的历史贡献**：1992年英镑危机后力挽狂澜建立英国通胀目标制体系，发明了经典的宏观预测工具“扇形图”（fan charts）。 ### `30:00` 央行沟通的演变：从“扇形图”的狂热到默文·金与沃什的反思 亚当·波森（Adam Posen）回顾了央行透明度与预测工具的历史演变。在英格兰银行前行长默文·金（Mervyn King）主政期间，英国央行大力推行著名的“扇形图”（fan charts）——一种用深浅渐变色块呈现经济预测概率区间的可视化工具。波森打趣道，虽然自己当时极度推崇这种做法，但这在央行界绝对算是个“小众产品”。伯南克（Ben Bernanke）等人在合著的著作中也曾大篇幅赞誉这一工具，认为它完美兼顾了透明度与问责制：央行不再假装拥有精准的水晶球，而是向公众坦承“这是我们的基准预测，但它有概率偏高或偏低，实际结果大概率落在这个区间内”。这种哲学坚信信息越多越好、定期且按时发布预测更有助于建立信誉，并迫使决策委员会在改变想法或吸取教训时公开认错。 然而，情况在默文·金退休后发生了剧烈转变。正如当前美联储候任主席凯文·沃什（Kevin Warsh）所主张的那样，默文·金近年来的发声越发尖锐，直指央行发布了太多噪音：“央行根本不擅长预测，既然我们预测得很烂，就不该抛出大量带有误导性的信息。”默文·金甚至更早就开始担忧，金融市场过度依赖央行的一言一行与前瞻指引，不仅导致市场不再向央行反向提供真实的定价信号，更滋生了严重的道德风险——投资者不再认真为风险定价。在沃什与默文·金看来，收益率本身以及围绕收益率的波动性本应是极其关键的政策信号；收益率波动适度加大并不是坏事，反而能倒逼市场敬畏风险、遏制资产泡沫，进而使市场价格更具信息价值。曾任职于纽约联储、美国财政部及贝莱德（BlackRock）高管的彼得·费舍尔（Peter Fisher）也公开表达过类似观点，对央行精准预测的能力持深度怀疑，并担忧央行虚假的笃定感会引发投机泡沫并蒙蔽市场信号。 谈及沃什在杰克逊霍尔（Jackson Hole）年会上的最新演讲，波森注意到沃什其实软化了此前的一些激进立场。例如，沃什在演讲中提到了“镜厅效应”（hall of mirrors），委婉承认了市场本身也并不总是对的——如果美联储一味盲目追逐市场的噪音，就会陷入怪异的恶性循环。波森对此表示赞赏，认为这远比沃什过去那句“盯球别盯裁判”（play the ball, not the referee，把市场当成球、央行当裁判）的糟糕言论成熟得多，后者曾极度激怒央行圈子。不过波森研判，沃什骨子里的本能依然倾向于“央行算不准未来、市场掌握更多信息、央行绝不能给市场‘喂饭’”。因此，美联储即将成立的沟通任务小组（communications committee）很可能会带来颠覆性的变革，彻底打破现行点阵图和过多图表的前瞻模式。 - **扇形图（Fan Charts）的核心逻辑**：将单一点位预测转化为基于概率的带状区间，旨在传递“概率性决策”而非绝对精准的预期。 - **“镜厅效应”的反思**：央行如果完全迎合市场预期定价，就会形成相互反射的噪音闭环，导致货币政策丧失定力。 - **关于市场纪律的观点分歧**：鹰派与市场派（如沃什、默文·金、费舍尔）主张容忍更高的收益率波动，以促使金融市场恢复自我定价与约束机制。 --- ### `33:30` 经济学家“投奔”AI巨头：科技奇点诱惑与独立发声的代价 讨论随即从高深的货币政策理论转向了当下知识界的现实裂变：大量来自顶尖大学和智库的社会科学家、尤其是资深经济学家，正以前所未有的规模出走，成群结队地加盟两家头部 AI 科技巨头。面对主持人关于这究竟是一场“吸积漩涡”（vortex）还是“人才流失”（brain drain）的提问，身为彼得森国际经济研究所（PIIE）所长、日常核心职责正是招募与留住顶尖学者的波森给出了极为深刻的剖析。 ```mermaid flowchart TD accTitle: 经济学家投身产业界的历史对比 accDescr: 对比互联网初期与当前AI浪潮中经济学家的角色转变、动机及潜在代价。 subgraph InternetEra["互联网繁荣期（Google / Amazon）"] A["内部机制设计"] --> B["拍卖与定价算法（如 Hal Varian）"] B --> C["边缘性产业政策游说"] end subgraph AIEra["当前生成式 AI 浪潮（如 Anthropic 等）"] D["巨额薪酬与技术狂热"] --> E["参与世纪级变革的核心叙事"] E --> F["直接设计 AI 转型与监管政策"] end F --> G{"学术独立性危机"} G --> H["公众对其研究结论天然打折"] G --> I["进入企业层级导致自我审查"] ``` 波森指出，这种学术人才涌向新兴产业的现象并非毫无先例。在早期的互联网繁荣时代，亚马逊（Amazon）和谷歌（Google）就曾招揽大批经济学家，例如麻省理工学院著名的微观经济学家哈尔·范里安（Hal Varian）出任谷歌首任首席经济学家。但彼时经济学家的工作主要集中在企业内部的生产力与商业机制构建上，例如设计广告拍卖系统或动态定价模型；几十年前基因工程和生物科技（如 Genentech）崛起时，诺贝尔奖得主与博士生也曾蜂拥下海创办初创公司。 然而，眼下的 AI 浪潮在诱惑力上比过去“高出了一个数量级”。除了令人瞠目结舌的高额薪酬和“与全宇宙最酷的人共事”的光环外，许多学者真诚地相信 AI 是“整整一个世纪以来最具颠覆性的技术”，并怀揣着投身其中改变世界的理想主义情怀。更特殊的是，以 Anthropic 等公司为例，它们招聘经济学家的公开目的之一，就是构想和制定指引全社会 AI 转型的公共政策。这使得局势变得格外微妙与尴尬：过去经济学家为亚马逊设计定价机制，哪怕被外界质疑，其技术属性仍很纯粹；但如今，当一位享誉全球的创新或增长经济学泰斗进入 AI 公司内部，随后向外界大声疾呼“AI 创造的工作岗位将是其摧毁的数倍”、“技术奇点已近”、“绝不能过早实施严厉监管”，无论其内心信仰多么真诚，公众与政策制定者都不可避免地会对其独立性和客观性大打折扣。 波森透露，自己近期曾与一位深受各大 AI 巨头猛烈招募、但最终坚决拒绝加盟的顶尖学者长谈。这位学者本身是一个坚定的“技术乐观派”，他之所以拒绝，正是意识到一旦迈入科技巨头的企业层级体系，就必然会失去完全独立的学者话语权，甚至会在研究选题上主动避嫌或自我审查。波森强调，虽然这种现象与赤裸裸的政治腐败截然不同，但依然令人深感不安。像彼得森研究所这样的独立智库和坚守学术阵地的学者至关重要，哪怕他们只能拿着六位数出头（low six figures）的薪水，也必须守护住不受资本俘获的第三方客观视角。 ### `40:00` AI 对就业与生产率的现实冲击：数据为何未现岗位流失？ 特蕾西·艾洛维（Tracy Alloway）将话题引向人工智能（AI）。就眼下的经济影响而言，外界看到的主要是对物价、通胀和资本开支（Capex）等短期指标的作用，而 AI 对全要素生产率乃至劳动就业市场的重塑，似乎仍停留在遥远的未来和高度理论化的推演中。针对“眼下是否已能在劳动力市场或生产率中看到切实证据”的提问，亚当·波森（Adam Posen）直言，相较于就业市场的实际变化，目前在生产率改善方面看到的指示性证据其实更为明确。 波森列举了彼得森国际经济研究所（PIIE）内外的顶级学者，包括马丁·乔赞帕（Martin Chorzempa）、杰德·科尔科（Jed Kolko，曾任拜登政府商务部首席经济学家、Indeed 首席经济学家）以及今年请假前往 AI 独角兽企业 Anthropic 任职的安东·科里内克（Anton Korinek）。从科尔科等劳动力市场专家的实证研究以及经济创新组织（Economic Innovation Group, EIG）的跟踪来看，数据中根本没有出现所谓的大规模失业潮。即便挑选两类此前公认最容易“被淘汰”的典型岗位——长途卡车司机和初级程序员，行业内的数据显示招聘依然在持续增长。 对于为何没有出现岗位替代，斯坦福大学的埃里克·布林约尔松（Erik Brynjolfsson）以及伦敦政治经济学院（LSE）的路易斯·加里卡诺（Luis Garicano）等学者提出了极具洞察力的理论： - **AI 部署的 J 曲线与技术吸收期**：布林约尔松提出技术存在“J 曲线效应”。企业从接触新技术到重塑业务流程、开发定制化应用并整合机器人等硬件，往往需要一到五年（可能不到十年，但接近五年）。在此期间，AI 正处于学习进化阶段，人机协同（Human-in-the-loop）反而能实现最高效率，因此短期内不会大量淘汰岗位。 - **“混乱工作”（Messy Jobs）理论**：加里卡诺等人指出，外界往往喜欢拿极端简单的岗位做模型，但绝大多数现实工作都高度依赖特定情境知识（Specific Knowledge）和复杂的社会协作网络，具有极高的互补性与专业化深度。咨询公司和国际组织发布的“最脆弱行业排行榜”往往严重误导公众。这类似于 19 世纪英国工业革命时期，真正被自动织布机取代的仅仅是从事纺织的狭窄技术工匠群体，其他更广泛职业的分工演变则远比事前预期的复杂得多。 ```mermaid flowchart TD accTitle: AI技术渗透与劳动力市场适应机制 accDescr: 展示技术引入到业务重构期间人机协同的过渡阶段 subgraph TechAdoption["技术重塑阶段（1~5年）"] A["生成式 AI 技术爆发"] --> B["业务流程与架构重塑"] B --> C["垂直应用开发与硬件整合"] end subgraph LaborImpact["就业市场应对"] C --> D["人机高效协同<br>（未见大规模裁员）"] D --> E["混乱工作特征：<br>深植情境知识与人际信任"] D --> F["边际冲击现形：<br>年轻求职者初级岗位招聘放缓"] end ``` ### `45:00` 生产率争论、人类交互溢价与广告插播 波森进一步分析，若说 AI 已经对就业市场产生了什么微弱的涟漪，那主要体现在对年轻、初入职场者的招聘有所放缓。然而，美联储前理事凯文·沃什（Kevin Warsh）在杰克逊霍尔演讲中的一句随口提及也是成立的：疫情之后各行各业经历了巨大的人员洗牌和岗位重新定义，不能想当然地把所有的招聘疲软都扣在 AI 头上。 至于生产率的增长到底有多少归功于 AI，学界目前存在非常健康的专业争论，这并非出于意识形态分歧，而是因为样本数据极其有限且模糊。波森自认对 AI 生产率的贡献倾向于“偏保守/悲观”，他认为过去一年半左右 AI 对增长的提振才刚刚显现，那些宣扬爆炸式超额增长的极端技术乐观派大多根本不是经济学家（乔·维森塔尔随口笑称“你是在说我吗？”，波森则幽默回应“我绝不会这么说，乔是全能通才”）。波森打趣引用祖母的老话：“要是真有那么好的事，我们可就太走运了，但现实还远未达标。” 乔·维森塔尔（Joe Weisenthal）补充支持了“混乱工作”假说。即便是卡车司机，日常也要处理仓库对接、突发调度等极其繁琐的人际与现场问题；EIG 的亚当·奥齐梅克（Adam Ozimek）也曾举例，即便现在有了自动演奏钢琴和流媒体，派对上依然会花钱雇真人演奏钢琴——因为真人琴师能观察现场宾客的情绪、看清场子冷热随时切歌提振节奏，这是机器替代不了的。波森随即分享了一则私下内幕趣闻： - 在一场不对外公开的高层闭门会议中，某位并非经济学家的 AI 圈超级顶流大佬正在大谈特谈。 - 一位知名经济学教授当场发问：“你之前一直跟全世界宣称‘就业世界末日’（Job Apocalypse）即将来临，但至今什么都没发生，为什么？” - 该 AI 大佬思索后答道：“事实证明，人类其实非常喜欢和真人打交道，并不总想和冰冷的机器互动。” - 波森调侃道，即便在向来被嘲讽“情商不高”的经济学家圈子里，大家当时的反应也是不约而同的“废话，这还用你说（Duh）”。就像大家不需要现在的生成式 AI 也能用 Spotify 听保罗·西蒙（Paul Simon）或比利·乔尔（Billy Joel）一样，人际互动的复杂温度是不可剥离的。 音频最后插播了一段商业赞助广告，宣传 OpenAI 推出的 ChatGPT Work 模式（工作模式）。广告提到部分人仅把 ChatGPT 当作智能搜索引擎，而 ChatGPT Work 则旨在跨应用与文件采取行动，协助用户完成耗时的项目调研、整合分散材料与简报，并在数小时内将杂乱的初始思路转化为可供审阅的交付初稿。该功能支持 Plus 与 Pro 订阅计划。 ### `50:00` 广告播报与英伟达芯片产值漏计的争论 主持人插播了辛辛那提保险（Cincinnati Insurance）与企业协作软件 Asana 的赞助播客广告后，节目重回对经济数据统计口径的深度探讨。Joe Weisenthal 抛出一个当下极具争议的现实话题：市场流传一份报告指出，类似英伟达（NVIDIA）这类企业虽然在本土负责芯片研发与架构设计，但其实际晶圆代工主要依赖台积电等海外供应链，导致美国现行统计数据未能完全捕捉其实际经济贡献。该分析甚至推算，这部分海外代工导致的数据遗漏可能使美国 GDP 增速被低估了多达 0.3 个百分点。 对此，Joe 提出疑问：即便 0.3% 不算小数目，但在市场上英伟达的实际利润已人尽皆知，而美联储的政策目标也并非直接钉死 GDP 读数，这种测算偏差是否真正具有政策意义？彼得森国际经济研究所（PIIE）所长 Adam Posen 回应称，GDP 的本质是在国家边界范围内，度量付出了多少劳动与生产，从而换取了多少收入。经济学界通常使用国内生产总值（GDP）、国内生产总收入（GDI）以及最终国内需求（Gross Final Demand）等多重维度进行交叉验证与校准。就长期路径而言，这些指标在大趋势上高度同步，单个顶尖科技公司的分类偏差并不会颠覆宏观经济的基本轮廓。 ```mermaid flowchart LR accTitle: GDP Mismeasurement in Tech Supply Chains accDescr: 芯片设计与海外制造导致的统计归属偏差 subgraph US["美国本土 (U.S.)"] A["研发与设计 (如 NVIDIA)"] --> B["高额营收与员工薪酬"] end subgraph Overseas["海外供应链 (如中国台湾代工)"] C["物理制造与封装"] end A -. 复杂供应链与增加值核算.-> C B --> D{"GDP 统计归属"} C --> D D --> E["数据口径争议: GDP/GNP 划分偏差 (~0.3%)"] ``` - **GDP 与 GDI 的双向校准**：GDP 与 GDI 理应在理论上等价，科技公司在复杂分工下的会计归属争议虽不会彻底改写宏观经济走向，但若每年累计漏算达 0.3% 且集中在 AI 核心赛道，对生产率测算的结论将大相径庭。 - **价值链划分的学术探讨**：PIIE 的同事 Martin Chorzempa 以及曾供职于美联储的 Joe Gagnon 均撰文深入剖析过该问题，重点探讨在跨国复杂供应链中，如何准确厘清研发附加值、知识产权收入与最终产品进出口的界限。 --- ### `55:00` 技术革命下的度量困境与“效用”悖论 Adam Posen 进一步指出，在市场经济的现实运行中，无论官方统计数据如何核算，英伟达的真实业绩丝毫不受影响——芯片实打实地销售出去，研发工程师拿到高额薪水，巨额现金流分别流向股东分红、资本开支与员工酬劳。但他强调，技术革命历史上反复上演的核心争论在于：传统 GDP 能否衡量真正的技术价值？经济学家通常对此有两种解释：第一，GDP 本就只衡量“市场交换价值”（Market Value），哪怕消费者获得的实际满足感远超其支付的价格，未发生交换的部分也不归 GDP 负责；第二，统计学采用享乐定价法（Hedonic Indexing）针对产品品质提升进行平减校正，但这主要体现于通胀指标（如物价平减指数）的下调，而非名义产出的直接攀升。 随后，对话转向了由 AI 引发的个人微观效用与宏观生产率之间的脱节。Posen 举例说，当用户输入指令让 ChatGPT 或 Claude 代订飞往杰克逊霍尔最便宜的机票，并要求“别把我和婴儿排在一起”（Tracy Alloway 打趣插话道：“其实她是要求别坐在 Joe 旁边，不想四小时飞机上面面相觑”），这确实节省了个人时间、避免了订票网站的繁琐折磨，提供了极高的个人效用（Utils）。然而，这种生活便利并未转化为账面上的 GDP 增长。Posen 由此强调生产率的“J 曲线效应”：如同 20 世纪 90 年代英特尔与德州仪器不断推动摩尔定律那样，初期芯片制造对 GDP 的推动有限，真正的宏观生产率飞跃与抗通胀效应，只有在沃尔玛、麦当劳和 UPS 等传统各行各业深度重构其商业工作流后，才会大规模爆发。Tracy 也顺带打趣回忆道，高盛首席经济学家 Jan Hatzius 等人此前就曾发布报告，论证《侠盗猎车手》（Grand Theft Auto / GTA）新一代画质的飞跃被统计数据严重低估，甚至开玩笑说等《GTA 6》发售时必须再请他们来聊聊这笔“画质生产率”。 --- ### `57:30` 鲍威尔美联储的盲区与坚挺的通胀前景 访谈尾声，Tracy Alloway 询问起 Posen 对通胀前景的最新判断，提到他此前曾与投行 Lazard 首席执行官 Peter Orszag 联名发文，预测美国通胀率到年底将重返 4%。Posen 确认了这一预测：当时其团队判断核心 PCE 约为 4%，而 CPI 还会略高一些；后续实际发布的 CPI 数据确实突破了 4%，让不少乐观预期通胀稳步回落的分析师大跌眼镜。 Posen 详细复盘了他此前对鲍威尔领导的美联储所作批评的依据： 1. **劳动力市场韧性远超预期**：美联储及市场模型长期低估了美国就业市场的吸收能力与内在刚性。 2. **金融条件实际上并未收紧**：信贷渠道保持相对通畅，名义加息并未对流动性造成预期的惩罚性压制。 3. **财政政策显著扩张**：政府部门持续宽松的财政支出在实质上冲销了部分紧缩效果。 4. **通胀惯性深度积蓄**：通胀长期高位徘徊长达五六十个月之久，早已形成了顽固的内生动能。 5. **政策滞后效应（J 曲线）**：加征关税与紧缩移民政策的负面供给冲击并不是立竿见影的，其成本推升效应正随着时间推移逐步发酵。 Posen 直言，在这一脆弱的宏观脆弱面之下，任何新增的外生通胀冲击都会被系统性放大。正当他谈到地缘政治等不可预知的突发极端冲击时，访谈在一句关于美国总统决策意图的深刻评述中推向最后的高潮。 ### `01:00:00` 核心通胀的粘性与美联储加息路径的预测 亚当·波森（Adam Posen）指出，无论是霍尔木兹海峡局势动荡对能源市场带来的冲击，还是其他宏观扰动，通胀冲击的再度发生早已不可避免。尽管未来一两个月内整体通胀可能因能源价格波动而小幅回落零点几个百分点，但正如美联储主席凯文·沃什（Kevin Warsh）及多位联邦公开市场委员会（FOMC）委员所强调的，核心服务业通胀（非进口商品）依然展现出极强的粘性。沃什在演讲中引用的数据显示，无论是 3 个月、6 个月还是 12 个月的移动平均线均处于上行趋势，目前通胀率稳定在 3% 尾部的高位，并随时有上冲 4% 的势头。 基于这一基本面，波森明确给出了自己的加息预测：美联储势必会进一步加息。如果 9 月会议未采取行动，那么 12 月必将加息；即使 9 月如期加息，12 月依然存在再次加息的必要。在接下来的 6 个月内，联邦基金利率预计将比当前水平提高 75 到 100 个基点。在此之前，通胀率很难真正如公众期望的那样大幅回落，而是将在 3.5% 到 4.5% 的区间内震荡徘徊，且面临显著的上行风险。在阐明这些判断后，主持人对波森在此次杰克逊霍尔会议现场的深入分享致谢，波森也表示自己一直是《Odd Lots》节目的忠实听众。 - 核心通胀粘性主要集中在剔除进口商品的服务类项目，多周期移动平均线呈上行趋势。 - 预计美联储将在未来半年内累计加息 75 至 100 个基点。 - 阶段性通胀预测区间维持在 3.5% 至 4.5%，面临持续的上行风险。 --- ### `01:01:45` 鲍威尔的历史声誉与公众对通胀的极度厌恶 在送别波森后，主持人乔·维森塔尔（Joe Weisenthal）与特雷西·阿洛威（Tracy Alloway）就美联储领导层的历史声誉展开了复盘讨论。维森塔尔提出一个耐人寻味的问题：未来的历史将如何评价杰罗姆·鲍威尔（Jerome Powell）的政治遗产？正如沃什所提及的，在鲍威尔任内，通胀率高于美联储官方目标的周期长达 64 至 65 个月。回顾历任美联储主席，阿瑟·伯恩斯（Arthur Burns）因任内高通胀而被普遍定性为表现糟糕；相比之下，本·伯南克（Ben Bernanke）即便经历了毁灭性的次贷危机以及长达数年的高失业率，如今公众和学界对他的评价却相对温厚。 这揭示了中央银行声誉评价中极其倾斜的标准：公众对失业率高企的宽容度，往往明显高于对物价飞涨的容忍度——人们对高通胀有着深入骨髓的痛恨。两人进一步探讨了 FOMC 内部的机制动向，例如鲍威尔治下委员会是否形成了某种结构性阻碍反对意见表达的惯性，以至于政策制定未能及早转向。此外，节目中提到的另一潜规则也引起了他们的共鸣：在美联储决策会议上，似乎存在一种不成文的默契，即单次会议上的异议票（dissents）通常被严格限制在极少数人之内，绝不允许出现大规模公开分歧。 - 鲍威尔任期内经历了长达 64 至 65 个月的高于目标通胀期，其声誉面临历史重估。 - 历史声誉对比：伯恩斯因通胀深陷负面评价，而伯南克在经历金融危机与高失业后依然保有正面评价，显示公众对通胀的反感远超失业。 - FOMC 内部存在非正式的行为规范，通常将单次决议中的公开异议人数控制在极低水平。 --- ### `01:03:08` 施工场景下的 AI 局限性与沃什的“偏鹰派”修辞 谈及上一部分波森提出的“凌乱工作（messy jobs）”理论，特雷西分享了一段亲身经历，以此反思当前对生成式人工智能的盲目崇拜。当下不少人鼓吹像承包商或蓝领技工这类工作很容易被 ChatGPT 替代，认为普通人只要向 AI 提问就能搞定一切房屋修缮。然而，特雷西与丈夫在给自家室外棚屋铺设屋顶时，听从了 ChatGPT 和 YouTube 教程的建议购买了一种固定瓦片的粘合剂，直到施工开始后才赫然发现该产品必须在华氏 60 度以上的温度下才能固化。当时的他们身处新英格兰（康涅狄格州），而聊天模型和算法视频根本无法识别地域气候的巨大差异，误套用了佛罗里达等温暖地区的操作标准。维森塔尔对此表示赞同，认为尽管 AI 模型表现惊艳，但在人类没有言明的隐性常识与情境判断力面前，AI 仍有难以逾越的鸿沟。 紧接着，两人对沃什在杰克逊霍尔的演讲风格进行了精辟拆解。维森塔尔用带有连字符的“偏鹰派（Hawk-ish）”来定义沃什的态度：一方面，沃什在演讲中明确强调了两个硬核事实——“通胀正朝着错误的方向发展”以及“现有货币政策限制性不足”，这两点在逻辑上直接导向了没有说出口的第三句话——“因此必须加息”；但另一方面，沃什的发言并没有表现出 2022 年鲍威尔那种“不惜一切代价砸死通胀、不见成效绝不收兵”的决绝姿态。特雷西对此提出质疑：如果沃什持有上述逻辑，为何在 7 月会议上没有果断加息？沃什当时声称需要等待更多数据，然而现实环境中经济数据源源不断，各类外部冲击更是接踵而至，这让人不得不怀疑美联储内部真正触发加息决策的门槛究竟有多高。 ```mermaid flowchart TD accTitle: Kevin Warsh 演讲逻辑推导 accDescr: 展示沃什演讲中明确表达的前提与隐含的加息结论之间的推导逻辑 subgraph Explicit["沃什演讲公开阐述的核心论点"] A["事实论断一：<br/>通胀正朝着错误方向发展"] B["政策评估二：<br/>当前货币政策的限制性不足"] end subgraph Implied["推导逻辑与市场留白"] C["未言明的必然结论：<br/>美联储必须加息"] D["决策悖论：<br/>7月未加息且强调等待更多数据"] end A --> C B --> C C -.-> D ``` - 特雷西 DIY 翻修棚顶的实际案例表明，AI 缺乏结合具体地理与环境参数（如新英格兰气候温标限制）的隐性判断力。 - 沃什的演讲结构：通过“通胀方向错误”与“政策限制性不足”两个既定前提，隐晦表达了加息诉求，但措辞留有余地，未达到 2022 年鲍威尔式的极端鹰派决心。 - 市场与分析人士对沃什政策阈值的疑虑：在宏观冲击常态化的背景下，等待“更多数据”往往成为拖延必要加息行动的托词。 ## Notable quotes > `03:45` 如果六个月后回看，能确认本次演讲是一个真正的政策拐点——即沃什前几个月的混乱表态仅仅是新官上任的“成长阵痛”，随后他被残酷的现实狠狠敲醒并重回正轨——那么这篇演讲完全可以上调至“B+”。 — Adam Posen > `05:00` 没有速度约束的方向毫无意义，这实质上让美联储退回到了过去四年的模糊状态；从听证会、首批新闻发布会、欧洲央行辛特拉会议到本次演讲，沃什最一贯的做法就是拒绝事前承诺任何具体指标，执意保留在每次议息会议前最后一刻随性做决定的特权。 — Adam Posen > `10:00` 如果在1999年，年事已高的格林斯潘不幸在美联储网球场上突发心脏病去世，美联储的所有信誉都会瞬间化为乌有，因为外界的信任完全绑架在他个人身上。 — Adam Posen > `12:08` 下一次会议就是我的最后一次会议。 — Paul Volcker > `20:00` 央行官员对财政部的真正职责，是协助财政部及其继任者在长期内可持续地发售美国国债。这意味着央行绝不能通过眼前的操纵、通胀化手段或政策“作弊”来配合短期财政需求；捍卫美国债务的长期信誉，必然要求央行行长有底气顶住行政与财政部门的施压。 — Adam Posen > `24:30` 我认为沃什是在向外界传递‘我接受更多异议与辩论’的信号，但坦白说，我不相信他。 — Adam Posen > `24:30` QE确实争议极大，大家可以按自己的意愿自由投票；但一旦表决通过开始执行，我们能否达成暂时的休战协议，不要在公开场合不断质疑QE到底有没有用？ — Mervyn King > `30:00` 央行根本不擅长预测，既然我们预测得很烂，就不该抛出大量带有误导性的信息。 — Mervyn King > `45:00` 要是真有那么好的事，我们可就太走运了，但现实还远未达标。 — Adam Posen > `45:00` 事实证明，人类其实非常喜欢和真人打交道，并不总想和冰冷的机器互动。 — AI 圈大佬 > `01:00:00` 如果 9 月会议未采取行动，那么 12 月必将加息；即使 9 月如期加息，12 月依然存在再次加息的必要。在接下来的 6 个月内，联邦基金利率预计将比当前水平提高 75 到 100 个基点。 — Adam Posen ## Mentioned Adam Posen, Kevin Warsh, Joe Weisenthal, Tracy Alloway, Jerome Powell, Alan Greenspan, Janet Yellen, Paul Volcker, Ben Bernanke, Rick Mishkin, Thomas Laubach, John Taylor, Donald Trump, Lisa Cook, Larry Meyer, Don Kohn, Austan Goolsbee, Mervyn King, Peter Fisher, Raghuram Rajan, Jeremy Stein, Karen Dynan, Hal Varian, Martin Chorzempa, Jed Kolko, Anton Korinek, Erik Brynjolfsson, Luis Garicano, Michael Strain, Diane Swonk, Adam Ozimek, Joe Gagnon, Peter Orszag, Arthur Burns, Federal Reserve, FOMC, Peterson Institute for International Economics (PIIE), Bank of England (BOE), European Central Bank (ECB), Hoover Institution, Anthropic, Google, Amazon, NVIDIA, TSMC, OpenAI, ChatGPT, Economic Innovation Group (EIG), KPMG, BlackRock, Asana, Cincinnati Insurance ## Video description <p>Kevin Warsh gave a hawkish speech at this year's annual Kansas City Fed Symposium in Jackson Hole, but that doesn't mean the challenges are over. Will the Fed actually pull the trigger on rate hikes? What happens if the central bank doesn't act quickly enough to curb inflation? And how exactly will the new chair want to leave his mark on the Fed? In this episode, we speak with Adam Posen, president of the Peterson Institute for International Economics, who gives us his take on the direction of monetary policy and he grades Warsh's speech. We also discuss why high inflation in the US is different than the inflation seen in other countries, how the Treasury's wading into the bond market affects Fed independence, which of the Fed task forces might be the most impactful, the "messy jobs" theory of AI, and what Warsh really means when he says he wants the FOMC to have a &ldquo;good family fight.&rdquo;<br><br>Get tickets to see <a href="https://events.bloombergevents.com/event/ODDLOTSLA/summary">Odd Lots live</a> in LA!</p><p>See <a href="https://omnystudio.com/listener">omnystudio.com/listener</a> for privacy information.</p> [Source](https://podtrac.com/pts/redirect.mp3/tracking.swap.fm/track/UVBrz8bN8aM2Xe47PEPu/traffic.omny.fm/d/clips/e73c998e-6e60-432f-8610-ae210140c5b1/8a94442e-5a74-4fa2-8b8d-ae27003a8d6b/ffabb0eb-48cc-4ad4-8c96-b4b701728958/audio.mp3?in_playlist=982f5071-765c-403d-969d-ae27003a8d83)

[3:45] Adam Posen: 如果六个月后回看，能确认本次演讲是一个真正的政策拐点——即沃什前几个月的混乱表态仅仅是新官上任的“成长阵痛”，随后他被残酷的现实狠狠敲醒并重回正轨——那么这篇演讲完全可以上调至“B+”。 [5:00] Adam Posen: 没有速度约束的方向毫无意义，这实质上让美联储退回到了过去四年的模糊状态；从听证会、首批新闻发布会、欧洲央行辛特拉会议到本次演讲，沃什最一贯的做法就是拒绝事前承诺任何具体指标，执意保留在每次议息会议前最后一刻随性做决定的特权。 [10:00] Adam Posen: 如果在1999年，年事已高的格林斯潘不幸在美联储网球场上突发心脏病去世，美联储的所有信誉都会瞬间化为乌有，因为外界的信任完全绑架在他个人身上。 [12:08] Paul Volcker: 下一次会议就是我的最后一次会议。 [20:00] Adam Posen: 央行官员对财政部的真正职责，是协助财政部及其继任者在长期内可持续地发售美国国债。这意味着央行绝不能通过眼前的操纵、通胀化手段或政策“作弊”来配合短期财政需求；捍卫美国债务的长期信誉，必然要求央行行长有底气顶住行政与财政部门的施压。 [24:30] Adam Posen: 我认为沃什是在向外界传递‘我接受更多异议与辩论’的信号，但坦白说，我不相信他。 [24:30] Mervyn King: QE确实争议极大，大家可以按自己的意愿自由投票；但一旦表决通过开始执行，我们能否达成暂时的休战协议，不要在公开场合不断质疑QE到底有没有用？ [30:00] Mervyn King: 央行根本不擅长预测，既然我们预测得很烂，就不该抛出大量带有误导性的信息。 [45:00] Adam Posen: 要是真有那么好的事，我们可就太走运了，但现实还远未达标。 [45:00] AI 圈大佬: 事实证明，人类其实非常喜欢和真人打交道，并不总想和冰冷的机器互动。 [60:00] Adam Posen: 如果 9 月会议未采取行动，那么 12 月必将加息；即使 9 月如期加息，12 月依然存在再次加息的必要。在接下来的 6 个月内，联邦基金利率预计将比当前水平提高 75 到 100 个基点。

[0:00] Joe Weisenthal: fiscal dominance [3:45] Adam Posen: growing pains [3:45] Adam Posen: smacked by reality [3:45] Adam Posen: out over his skis [6:35] Kevin Warsh: right direction and at the right speed [9:55] Adam Posen: Weekend at Bernie's problem [9:55] Kevin Warsh: good family fight [14:15] Kevin Warsh: the Fed's fault [14:15] Adam Posen: enemy of the state [28:30] Kevin Warsh: play the ball, not the referee [40:00] Luis Garicano: messy jobs [60:00] Adam Posen: for reals [64:15] Joe Weisenthal: hawk-ish with a dash

[0:00] Did you ever notice how you spend hours shopping online only to pause a checkout because you wonder if you trusted enough to hit buy now? Agentic Commerce is testing that moment more than ever. That's where PayPal comes in. With 25 years of checkouts, 400 million consumer accounts globally, and the benefit of fraud protection. So no matter where a purchase starts, it ends with trust. Built for payments, growth in Agentic. PayPal Open. Built for all business. Visit PayPalOpen.com. Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans Healthcare doesn always work great If you ever waited on a refill or couldn schedule an appointment you get it That the kind of stuff Optum is changing They're using data and technology to integrate patient care, pharmacy, and everything else. So healthcare is connected, not complicated. What's that look like? Cheaper prescriptions that are easier to get and care that looks at the whole person, how you need it. Optum is helping make healthcare work as one for everyone. Learn more at business.optum.com. Bloomberg Audio Studios, podcasts, radio, news. Hello and welcome to another episode of the Odd Lots podcast. I'm Jill Weisenthal. And I'm Tracy Allaway. Tracy, another Jackson Hole episode. You know, I think with central banking and some of these topics, people are pretty polite. You know, it's like sometimes you talk to people, it's a little bit hard to know what they're really thinking about all these topics, what they feel. Sometimes I do perceive a certain lack of I don want to be candor Let say people are very diplomatic People are very diplomatic Thank you for finding the word That a diplomatic answer to the point that you making about diplomacy at an official Fed function People are very diplomatic, and I'm pro-diplomacy, but sometimes it's nice to just check in with someone. Just like, all right, let's give it to us straight. I'm waiting for you to describe our next guest. No, I'm saying, all right, I'm just going to jump right into it. A guest who I believe when we talk to him, I feel like here's someone who's giving it to us straight, who's not as concerned perhaps with being overly diplomatic. Someone who's not afraid to utter the words fiscal dominance. Yeah, someone who's just afraid, someone who just tells it like it is. Anyway, very excited to say here in Jackson Hole, back with the perfect guest, someone we've had on the podcast multiple times before, Peterson Institute President Adam Posen, also formerly a member of the Bank of England's Monetary Policy Committee. so Adam thank you so much for coming back on Odd Lots after that intro thank you so much I don't know whether it's hype or warning you have to throw some fastballs but what did you think of this is the first interview we've done since the speech so what did you think of Sherman Warsh's speech even though no no no even though I knew that was the question that was coming You know if you were grading it it a B speech It a B speech by normal standards It much more positive because of the situation we were in He, the chair, well, you know, let's be blunt, as you said, the chair had created a huge amount of not just confusion, but in central banking circles in the sophisticated observers like you about what he really thinks. And this was both short term, does he want to raise rates? Does he not want to raise rates? But also medium long term, what is behind all this task forces and the rhetoric? And then he obviously at the July press conference totally messed things up. Now, to be fair, three of his four predecessors as chair had major stumbles in one of their early press conferences, Greenspan, Yellen, Powell. So in a sense, if we are able six months from now to look back and say this was a pivot point. And so the stuff that Kevin, excuse me, the chair was saying for the first few months was just growing pains. And then he got smacked by reality. Yeah. All good. then this speech gets upgraded to a B plus and we're all happy. [5:00] And I think there are a number of people now who hope that's true with some confidence more than before, because the big thing, and Tracy and I talked briefly about this before coming on, is how different this was than last year. Because last year was all about the attacks on the Fed and the fears for the independents and the standing ovation for Jay Powell was his last thing, well, last Jackson Hole as chair. And this year, I think deliberately, the chair and the team wanted to make it much more normal, and it passed much more normal. The second thing is, as a European central banker said to me at breakfast this morning, but I think it's widely shared sentiment. For all the anti-forward guidance, he basically said up they have to hike. Now, as you know, and we talked about a few months ago, I've been urging them to hike for a while. I think inflation is real. So, again, I'm glad if they're making that call. But yeah, at this point, they really guided. Because if you read, there were four sections to the speech. And the fourth section basically does the litany of the reasons you would think inflation is not going to stay stable or it going to persist and maybe go up from here even if it goes down for a month or two for a while And basically the next sentence is therefore we have a tight new guise Didn't say the next sentence. So things could get really messy because if they don't hike, then people start saying, was that because of Trump? Was the chair out over his skis, meaning the head of the committee? If the chair is the head of the committee, it's better than being behind the committee, but still worrisome. Then the third thing is he did clean up some small but important things. So like he had talked for a while about what's the right target, we don't really know what inflation is. And yesterday he stated very clearly, core PC, 2%, that's the target. Similarly, he said something I think quite useful about he doesn't think wage inflation is very good predictor of overall inflation, which is a perfectly reasonable position to take. And then later he said, I think financial conditions are important. I know you guys have talked a lot about this at the industry level. All of these were normal sound either mainstream or very defensible central bank things And so that would have been sort of except for the fact that we had these two months three months of weird stuff And so it takes on more credibility I don want to go on too long but two other points if I may So the other two takeaways for me from the speech were first he does this whole second section which is about his principles and if I were him, I would have made this, and I was surprised, I would have made this a much narrower, much more practical speech. I basically would have cut that section. It was a long speech for someone who says they don't like forward guidance. Exactly. No, you're absolutely right, Tracy. I was like, people ask me ahead of time what I thought was going to happen. I thought he was going to compete with Jay Powell from like four years ago for the shortest possible speech. I thought he was just going to, you know, he didn't. So I think that muddles his message because some of the principal stuff, it raises more questions than answers. It's either a bromide or what does he really mean by that? And I don't think that was useful. But, you know, that's not a big deal. The big deal that I'm worried about is, and a lot of ex-Fed, ex-Central Bank people, academics are worried about, is he seems to be still even through yesterday speech for all its improvements trying to maximize his last discretion Yeah So this is the thing I noticed because at the very end of the speech he talks about, well, inflation has to be heading in the right direction and at the right speed. And I don't know what the right speed is. The right speed sounds very subjective to me. Absolutely. And right direction doesn't mean anything unless you're saying the target. I mean, without the speed, just to emphasize your point, without the speed, then we're back with what we had the last four years, which is inflation's above target. Well, we know what the direction should be, but if we don't declare a speed, we're going to bring it back to target. It doesn't mean anything. So, yeah, I think it's really striking because that, to me, is the most consistent thing through the stuff he said from his hearings, his confirmation hearings, his first two press conferences, his remarks at ECB Sintra and then yesterday, is he's sort of reserving the right to make up his mind at the last minute, before every meeting, without pre-committing what things he's looking at. He gave us a little bit of that. I mean, to his credit, he did say some things about which indicators he likes better than others, and that's good. And you don't want him to be inflexible. But [10:00] It's really odd for somebody coming from the Hoover Institution, who's been seen as a conservative, who was mentored by the late John Taylor, but also just in a broader central banking context, everybody tries to situate themselves between what's called rules versus discretion. Is it like, I'm going to raise rates every time monetary growth goes up, or something like that, which generally is a bad idea if you're too strict, or pure discretion. I'm going to, you know, I don't have to justify. I'm just going to make the right call every time. And that's basically Greenspan circa 1999. And Greenspan could get away with it, partly because he was very good, and partly because by that point he had such dominance over the committee. And partly it caused trouble for later, because then you had the weekend at Ernie's problem, which is if Greenspan, God forbid, drops dead, and we obviously just lost him recently, but in 1999. I have the image in my head now. No, no. In 1999, you were worried if then already somewhat old Alan Greenspan had a heart attack on the Fed's tennis court, all the credibility went away because it was all about him Oh interesting And this is why Ben Bernanke Rick Mishkin Thomas Laubach and I pushed the inflation targeting idea And Ben very consciously when he was chair pushed accountability because he wanted it to be not just all about one person. And I think this to me is not talked about enough and is the most worrying thing about where Chair Warsh has gone. So this is exactly what my next question was going to be, you have been a central banker at the BOE, not necessarily the Fed, but you have insight into how these monetary policy making committees actually work. What is the role of the Fed chair in its current form as you envision it? Is Warsh supposed to be trying to get everyone on his side and get them voting in his direction? Or is he supposed to be, I don't know, synthesizing a common position from the committee? That's a really good question, Tracy. And in the Fed system, that's not specified. So, and it varies over time, partly because of the chairman, chair, excuse me, partly because of the chair, partly because of the political surroundings, partly because of the economy. But for example, at the European Central Bank, it is for all the attention deservedly President Lagarde gets it is very much meant to be a consensus organization and really try to get as many people on board as possible and that probably because they representing nations not districts within one nation. At the Bank of England, when I served, and still, and this is something Chair Warsh invokes now, invokes, I should say, is it's all about discussion, what Kevin calls friendly debate or whatever. Family fight. Yeah, that's what he says. You're right. And the Bank of England really prioritizes that. The Fed has generally, on average, given more power and deference to the chair, certainly basically since Volcker, so the last 45 years. It varies over time, but essentially there's always something. So a financial crisis. Ben Bernanke started running the committee trying to open it up more, more debate, but financial crisis came and you really wanted Ben making the decisions and there wasn't time and then there were these crazy US rules about you can't have more than four people in the room from the committee at one time without it becoming a public. Then you have to take notes and whatever and so like ended up it would be the New York Fed president Chair Bernanke Kevin Warsh and Don Cohn were the four people in the room And that was the right call given the situation So anyway, this is part of the reason I worry about the discretion, because the ethic of the committee and the norm of the committee in the Fed system is you don't contradict the chair. You might dissent, but even there, as was pointed out by the former Fed Governor Larry Meyer, a while back. There's sort of an informal rule. Only so many people dissent on the committee at any one meeting, and the chair never loses a vote. Even Volcker, once it was clear he was going to lose a vote, he basically said, next meeting is my last meeting. That's super interesting. Well, let's actually, so one of the comments Chairman Worscht said, he basically said, I don't know, 64, 65 months of above-target inflation. That's true. But he also said, and this is the Fed's fault. And so, you know, obviously, that means it's in large part, he claims his predecessor's fault. But it also occurs to me, you know, inflation has been above target around the world in many respects, and it's not just a U.S. problem. I'm curious. [15:00] That comment, like, is it seems kind of true? Like, is that is that fair? Is that correct? I think it's fair. I mean, I think what was unfair in the run up to his appointment by President Trump and in some of the initial statements, there was a lot of talk about diplomatic, not diplomatic. There was a lot of pretty nasty sounding tone about the previous regime and, you know, I'm going to reform, I'm going to clean house on my watch. That was unnecessary. But I do think it is entirely justified to say, CarePal, with the deference and buy-in of the committee, basically was late to the game in hiking in 2022, was early, if not, in my view, profoundly mistaken to cut multiple times last year, and is behind the curve again. And the international comparison actually doesn't flatter the U.S. You have to do a little bit of careful things on the data. But basically if you control for three things which is how dependent on imported energy are you What was the inflation going into the last few years How important is the energy Bank of England's almost the same as the Fed. But additionally, this gets into the independent stuff, because part of the reason the U.S. has more inflation, in my view, in many people's view, is because Biden did do a fiscal blowout on coming into office that probably wasn't necessary. And then there's been no consolidation. And obviously Trump in his first year did this long-term set of tax cuts. and so but again if you take Kevin's what you quoted him you know it's ultimately up to the Fed if you take that seriously then all this fiscal laxity in discipline should be another reason to hike so anyway much too long but I think even a sober temperate thing suggests no Powell did get it wrong Now it not he got it horrendously wrong It's not he got it wrong for the wrong reasons. But, you know, when you and I talked, the three of us talked about this last year, which is in the year before, that at Jackson Hole, there was this drumbeat from the Fed staff and members. The labor markets could be really weak. The inflation is going down. And there were people like me or Michael Strain, AEI, and Diane Swank in KPMG who were out there saying, we don't see this. Now, I'm not saying necessarily they had to listen to us and could, and it didn't. But just to say they actually did get something wrong. There were people who were saying, no, the labor market is not going to collapse in the next two years. And they kept saying, yes, it will. So, yeah, they did get some things wrong. What's the sort of bar for political interference with the Fed? Because, you know, it could be anything from just the president talking about how he thinks rates are artificially high. It could be maybe something a little bit different, like the fiscal dominance idea to the extent that the Fed now feels hamstrung in some way in terms of what it can do with rates. Again you right to raise it Tracy I mean so the first thing to say is a certain amount of yelling at or scapegoating the Fed is part of the game And in fact, in previous presidencies, it kind of sometimes was healthy, right? So it's like, oh God, Fed has to raise rates to deal with inflation. I don't want to be blamed for it. So I'm going to go out there and yell, but meanwhile, in the background, wink, wink, Fed, do your job. And that was under Volcker, under Greenspan, under Bernanke, under Yellen, I mean, that was always understood. So a certain amount of, not like President Trump tweeting, the Fed chair is an enemy of the state, not using lawfare to go after Lisa Cook, Governor Cook, or now Governor Powell. I mean, that's outrageous and arguably criminal. But a certain amount of carping about the central bank, that's fine. Where the rubber hits the road is partly what you said, which is when does the government treasury secretary or president who's trying to sell U.S. bonds say, you have to help me sell U.S. bonds, whether or not inflation is high. And in that case, what a central banker should do, a Fed president or a Fed governor should do is say, [20:00] I'm here to help you sell U.S. bonds and your successors sell U.S. bonds and your successors sell U.S. bonds. And that means I don't do cheating and manipulation now. I'm here to help you sell the bonds. You're preserving the long-term credibility of U.S. debt. And so you have to be willing to stand up to that. And that gets us into the potential really big situation now. But the second thing, and this is the part that was really unprecedented under Trump besides the individual illegal attacks on individual members, is they were, and this was very loud last year, they were threatening to take away the votes of some of the Reserve Bank presidents or quickly turn over the Reserve Bank presidents, not get rid of them and put in their own political appointees rather than having the staggered terms as intended. and there was talk about changing the Fed's mandate and then the things about the Fed's budget. The big one, economically speaking, is the caving in in the face of fiscal pressure. But almost as big and as a means to that one is the threats to politicize turn over fire remove the functional independence of the Fed by messing with who's appointed when. Did you ever notice how you spend hours shopping online only to pause right before you hit buy now? That tiny hesitation, the one where you wonder if it is trustworthy, can make or break the sale. Now, with AI changing the way consumers discover and compare things, that split-second trust question matters more than ever. That's where PayPal comes in. For more than 25 years, PayPal has been a leader in online payments. And now they're at the forefront of agentic commerce, making it work for businesses. Letting them maintain control of their brand and their customer relationships. So even as the way people shop changes, the moment that matters most still feels familiar and deeply dependable. Built for payments, growth, and agentic. PayPal Open Built for all business Visit PayPalOpen to get started Some people treat ChatGPT like some kind of smart search engine And some use it to get started grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting work mode, available on plus and pro plans. If you listen to financial news, you know a lot of time is spent thinking about what's next, the next opportunity, the next investment, the next move. But sometimes what matters most is being ready for what you never saw coming. A storm damages a second home. A pipe bursts above a collection you spent decades building. An unexpected loss puts a growing business on pause. Those are the days no one sees coming. But for more than 75 years, Cincinnati Insurance has been ready when they do With deep expertise coverage for businesses homes valuables and more Cincinnati Insurance works with independent agents who take the time to understand what matters to you Together, they help protect more than what you own. They help protect what you're building toward. And when a bad day does come, Cincinnati Insurance has real people there to help make things right. Because planning for the future isn't only about knowing what's next, it's about making sure you're ready for what you can't predict. Let Cincinnati insurance make your bad day better. Find an independent agent at CINFIN.com. I'm just thinking about a few things that have been said. Last year at this conference, I asked Chicago Fed President Austin Gillespie, why are dissents rare at the Fed? One of his answers is that Chairman Powell was good at building consensus. But now I'm sort of curious, was there a perception under the power of the Fed? [25:00] years that he was too good at building consensus? Did members of the FOMC feel that actually that meeting room was not a real site for debate? And when Kevin Warsh says, we want the family fight, is he saying, no, we're, I'm okay with, unlike my predecessor, I'm actually okay if you don't agree with me? I think he's saying, I think Kevin Warsh is saying what you just said, Joe, that I am okay with more dissent and more debate. I don't believe him. Okay. But if you take it at face value, he's right. And that's what I meant in part. You think he's right? He's right that debate had gotten too little, and it's healthy to have debate. I think Powell, it's partly he was very good inside the building, but I think it was two other things. I think he smartly and justifiably first under COVID, and then when the Fed was under such overt attack by Trump saying let's not dissent and create divisions in the committee casually. Let's make the bar very high so if you dissent I understand but let not show division because that just gives the attackers on the Fed or in the case of COVID that just gives the sense of panic And there is always this temptation So like when I was on the Bank of England MPC which was 2009 to 12 so I was there during the financial crisis And there was a point at which the governor and Bank of England had a much more live debate, publicly at least, than the Fed. then Governor Mervyn King came at one point and said look, quantitative easing is really controversial we're going to take the vote, you'll vote however you want but once we've done it, can we have a moratorium on people asking whether or not quantitative easing works or not for a while, and I totally supported that I mean, again, it's an emergency, you don't want to just oh well, you know, according to Vixell, I'm not sure You know, the usual central bank thing. I'm giving a talk in Glasgow and I just sort of wander off into monetary theory. So I think not through bad motivations but I think Powell ended up between the COVID panic and then the rallying solidarity and people not wanting to look like they were undermining the chairman while the Fed was under attack You did end up getting much lower than usual descents And then what happens which shouldn but does happen is it builds momentum So if you gone 12 meetings 15 meetings without anybody dissenting, the bar psychologically becomes higher for someone to dissent. So again, I hope I'm wrong. I don't think Chair Warsh really means it. I hope he does. But taking it literally, yeah, we went through a period where there was too little dissent. So you mentioned Mervyn King just then, and King is heading the new task force on Fed communication. So I'm curious whether you have any insights or perhaps informed guesses about what the problem is that Warsh slash King are now trying to solve? Well, I know well two of the three members of that committee. So I worked with Governor King. He was governor, and I knew him before that when I was at the Bank of England. and Peter Fisher, the former New York Fed and U.S. Treasury senior official, is on the board of directors of the Pearson Institute. So just full disclosure. So I'm not talking to them directly about confidential things, but just disclosure. I think this is the committee that has the biggest chance of a surprise to being radical I think people are very concentrated on the balance sheet committee and I think actually they likely to come out with much more sober practical, smaller scale recommendations than people think they are. That's just my guess. Yeah. But knowing who's on that committee, Raghu Rajan, Jennifer Stein, Karen Dynan, who's also a colleague of mine, but I think the communications phone is going to surprise people because both Mervyn and Peter and I'm only referring to their public statements have gotten pretty radically skeptical about central bank communications in recent years so Mervyn King was actually the leader of the inflation targeting movement long before we wrote our book he's the one who in 1992 when the British pound fell out of the exchange rate mechanism and there's basically a crisis. We need to have something to anchor in the UK. He's the one who leapt into action with a couple key staffers. He was chief economist at the bank at the time and said, look, these little countries are trying inflation targeting, but we think it can work for us, and he did. And so he was the father or the parent of what used to be known as the fan charts. I love the fan charts. [30:00] I was the only one, but I really liked that. No, no, no. There were a few of us scattered around, but it was definitely a niche product. And there are problems that are worse with fan charts, and the Bank of England has changed it. But basically, Mervyn was leading the effort, and we cited this hugely. Bernanke et al. cited this hugely in our book, that they were, again, back to the combination of accountability and transparency, that you were sending out a forecast. You were being much more explicit than central banks usually were about the forecast. You were with the fan charts, which are essentially colored spreads about how likely a given outcome would be. Here's what we centrally think is going to happen. Here's what may happen. You were trying to give the message this was probabilistic, that we're making our best guess, but it's probably going to err on the high side and it's going to be somewhere around here. So anyway, it was very, very much about transparency, accountability, and more information is better. Regular information is better. Commitment to specific releases at specific times is better. So you have a track record and the committee has to admit when it changes its mind or learn something Anyway in more recent years since Governor King left the bank retired he's become quite loud in the spirit of some of the things Chair Warsh has been saying, that it's too much noise, central banks can't forecast, we're lousy at forecasting, given that we shouldn't be putting out all this information that's not really misleading, And Mervyn also is, from even before that, was very concerned about this idea of the markets getting too dependent on the information and the future plans of the central bank and then not giving as much information back and also having more moral hazard that they're not going to take the price and risk. This is the idea that yields are an important policy signal for central bankers, or they should be. They should be. And it should be a two-way street. And it should be yields and volatility around yields should be higher than it had been. This is their point of view, not mine. It should be higher than it had been because that forces markets to take risk more seriously and price things better. And so you have less not zero but you have less chance of a bubble and less inequality of things running out of them And then at the same time that should mean the pricing that the markets give you is more informative And Peter Fisher, who after he was at Fed New York and Treasury, was head of Asia and other senior roles at BlackRock, he feels very strongly these two points as well, at least what I know from his public statements. He's very skeptical about central banks' ability to forecast and very worried about central banks' certainty about central banks causing bubbles in discipline and then clouding market signals. So even though Chair Warsh in the speech here at Jackson Hole backed off a little bit of some of the more extreme statements of that, He admitted that he had this thing about a hall of mirrors in the speech, which is a way of admitting the market isn't always right. And if the Fed chases the market noise, you may get a weird dynamic. And I was glad he said that. Much better than play the ball, not the referee, which was garbage. That was the previous message was like, oh, market, you're the ball. Like, off you go. Yeah, that was really bad. But anyway of all the things that annoyed the central banking community that may have been the most annoying But anyway so just to say I think Chair Warsh true beliefs and instincts though are much more in the we can get any forecasts right The market has much more information on a lot of things than we do. We have a moral hazard issue of not spoon-feeding the markets. So I think the communications committee is probably where we're going to get the biggest surprise. Yeah, I find myself thinking maybe all these things, various charts, dots, et cetera, served a real purpose, maybe post-GFC, and maybe it's just a lot of noise right now. Since we're talking about this sort of monetary policy at the theoretical level, which is great, and all this, I'm just sort of maybe sort of seg a little bit here. I'm curious, you know, from Think Tank World, what do you think about so many social scientists, including a lot of economists, suddenly either leaving universities or leaving think tanks and joining one of two very big AI companies? When you think about the sort of future of social scientists, not so much like, well, [35:00] models, which I'm sure there's going to be tons of that. But just the sort of, depending on which side you're on, either vortex or brain drain of people of stature moving into these companies. What do you think? How does that make you feel? I mean, as someone whose main job is to hire, develop, and retain talent, I think about this a lot. And it is, as I think you applied, Joe, I mean, I think it's not unprecedented. So during the Internet boom, when Amazon was first coming up and Google was first coming up, they hired economists. They were hiring them more to be useful, like Amazon's pricing model, useful in terms of internal production. Design and auction system. Right. So like the very famous MIT microeconomist, Hal Varian, became Google's first chief economist and brought with him and hired a bunch of really smart people. But they also were there to talk about policy and help explain why the technology was good and what kinds of regulations they wanted and didn't want. And that's normal. I mean, that's okay. But there was also a little bit of this sense, which we're seeing now, an order of magnitude higher. Oh my God I can make real money Oh my God that where the cool people are Oh my God I can be part of changing the world So it very seductive And we seen this in other fields You know when genetic engineering had a boomlet many decades ago all these Nobel Prize winners and grad students would leave academic jobs and set up startups to, you know, genentech, things like that. right now I think it's challenging it's challenging because the case that can be made and I know I have colleagues I know who have gone in house at some of these companies the case that can be made that this is the most transformative technology of a century and it has the potential to do enormous good for the world is legit it may be overhyped, it may take longer to get there, we may have to care more about how many people we displace and what happens on way, but the idea that that is a valid view and you want to be part of that, I think is real, and so beyond the money and the sense of I want to be doing the cool stuff which matter because think tanks you pay well compared to average American salaries which you don pay like those people You know I understand it You can also go one step further, which is, unlike, say, Google or Amazon when they were coming up, where there were policy issues, but actually a lot of us didn't realize what all the policy issues were. and the economists who were hired, some of them worked on policy, but most of them were, like we said, working on sort of internal tools. Now, a lot of what the hiring, say, for example, Anthropic is doing, is by their own account about trying to envision what the right policies are to make the transition for AI work. And at least among the economists I know who've gone there, that's a sincere belief. but it does make it more awkward because it's one thing to say oh I work for Amazon I came up with the pricing auction model and that's why you get that and people say you're ripping me off and you say no I'm not but when it's like someone who's world class expert on the economics of innovation or of growth or whatever goes in and then starts saying well you know they going to create three times as many jobs as they destroy and the singularity is near and so you can regulate it right now because we on the cusp of society They may sincerely believe it, but people understandably discount their views once they're in-house and getting money. and also, and I was talking about this, a very noted academic who has been pursued by all these places who turned it down not to go. Hero. Well, and this is a person who's very techno-optimist. I mean, so it's not about that. But there's also just you do inherently, these are still companies. I mean, you're becoming part of a hierarchy. You're losing your independent voice. what you choose to work on you know you may you may choose not to work on that which you might have worked on because now you work for this company so again is it outright corruption of the sort we're seeing in the trump administration no but it is unsettling and i think there's a role for this academic colleague of mine the people at the peterson institute and like institutes to i hope be willing to settle for the low six-figure salaries which are [40:00] pretty darn good and influence policy without going on staff. Just on the AI side of things, so at the moment, if you look at the economic impact, it's very much sort of short-term effects on prices and inflation and capital investment and everything else, you know, the impact on productivity, even jobs is much further off into the future and very theoretical. Do you see any evidence of an impact on the labor market side or productivity here and now? Much more indicative evidence of improvement in productivity, Tracy, than on effects on the labor market. So speaking of colleagues of mine, Martin Trozempa and Jed Kolko are both top economists who work at Peterson and work on these issues. Anton Koronek is affiliated with Peterson, is on leave this year. he went to work at Anthropic this year. But anyway, if you look at their work, and in particular, I would cite the work of Jed Kolko. And he was chief economist at the U.S. Department of Commerce under Biden. He was at Indeed. Was it Indeed or ADP? Yeah, Indeed, I believe. Yeah So he a real practical labor market economist And he and there are a number of others There another think tank called Economic Innovation Group There are a bunch of them And you just not seeing the data So the number of hires, even of coders, if you had to pick the two jobs where you most thought they're toast, would be long-haul truckers and coders, lower-level coders. And we're just not seeing it. So job growth continues in those industries. I think there's been a lot of, yes, theory, but really important theory work done, not by my team, but by people like Eric Bjornolfson at Stanford or Luis Garricano at London School of Economics about why – those are just two examples, but they've done great work – about why we haven't seen the job displacement yet. and partly it's a matter of time, as you indicated. Time keeps coming up. It's how fast things happen. That there is arguably a period in which it is efficient to have the human and the AI working together, especially as the AI is in sort of learning mode, and especially as the businesses, just like with the Internet you have the technology and then you have to figure out how to reform your business to make use of it So there a period there we don know if it one year five years it probably less than 10 years it probably closer to five But there's a period there where the businesses are changing and the specialized applications are being created and things like robotics are being integrated. And during that period, we probably don't lose many jobs. Again, long-haul truckers are not trying to make fun of them. I mean, they may be the ones. And so, Brynjolfsson has spoken about a J-curve, that it takes time for this to happen. And it's similar to the argument I made with you all about the effect of tariffs and migration last time I was on, that businesses, people have to make decisions and implement things. It takes time. another argument which um louis garcano and his co-authors made is what they call messy jobs which is the idea that actually again you can you can try to come up with the uh sort of extreme case long-haul trucker job but most almost every job even arguably long-haul truckers um have a lot more specific knowledge and are embedded in a lot of relationships that there much more complementarity and much more I don want to say uniqueness but specialization than people appreciate And so just sort of doing a lot of these things that get published by consulting firms or international institutions, like these are the most exposed sectors, these are the most exposed sectors, probably are misleading. And we know this from the past, that again, in the Industrial Revolution in 19th century England, the analog to the long haul truck driver was the skilled artisan who did weaving. And so those people literally did get replaced by the automated weaving stuff. But if you had done this kind of study before the Industrial Revolution, outside of that very narrow, they're real people, but that very narrow job description, the displacement was very different than you would have expected. anyway so so the job stuff it's probably coming if it is coming it is showing up in the lack of hiring of younger people yeah and chair wars they had a throwaway line in a speech yesterday but it happens to be one i agree with and we again we've talked about this previously that a lot of the [45:00] The huge shifting around of people and redefining jobs after COVID and the reemployment. And so you can't automatically say the lack of hiring is AI. I mean, it may be contributing, but there's other stuff. Productivity is different. So there, I think there's a legitimate active debate. And it's not a debate ideologically or anything. It's because it's generally, it's hard. You got a limited amount of data and you're trying to figure out what's going on. I think there's a legitimate debate about how much productivity growth we've got that's due to AI. Is it the real big money stuff yet? How soon? I'm at probably just this side of pessimistic on that in the sense that I think going back to the jobs reshuffling after COVID, I don't think the AI productivity kicked in. the AI productivity boost to growth kicked in until very recently, last year and a half. But that is much more an open debate. Nobody except some really, really crazy techno-optimists who talk about staying in your lane, who are not economists. You mean Joe No no no no I would never say such a thing He a man for all seasons Finally someone says I very diplomatic They, you know, I think the extreme calls about, you know, huge amounts of productivity growth, as my grandmother used to say, we should be so lucky. Yeah. But we ain't there yet. I think, by the way, I find the messy jobs theory to be quite compelling. And by the way, the long-haul truckers would absolutely say, look, there's a lot more to this job. Oh, I agree. And include, you know, like all the things that they have to do with the warehouse. No, I didn't mean to mention that. But I think it illustrates that point. Let me tell you what's actually involved. Another one, you know, people point out, Adam Ozemeck has pointed out that people still get hired to play the piano at parties. Yeah, yeah. But, you know, as others would argue, well, the person responds to the crowd. They see what the mood is. These things that the player panel could not do, they could say, like, oh, maybe this crowd looks like it needs an uptempo thing. So I do think these jobs are much more probably complex. I'm glad to hear you say that. And again I was using the truckers just because that the example No no I know But I think it a good thing But I agree with you And Adam Ozimek who at EIG is one of the other people I think is doing great work on this and comes down in this sort of it may come yet, but it hasn't happened yet. But I think the really important point you said about the messy jobs is like with the piano. So I was in a meeting, supposed to be off the record. I won't go into details, but a very big, big, big, big, big shot from the AI community, not an economist. There are no economist big shots. from the AI community was talking and a pretty famous academic economist asked this person, well, you kept telling everyone there's going to be a job apocalypse. Hasn't happened. Why do you think that is? Just open question. And among other things, this person said, the big, big, big, big shot said, well, you know, it turns out people really like dealing with humans and don't always want to deal with machines. And even among the not exactly emotional IQ high economist community we're all kind of like, duh. You know, I mean, that's what you mean about the live piano player. I mean, I didn't need AI to have Spotify, I mean, current generation AI to have Spotify, you know, like, figure out if I like Paul Simon and Billy Joel then I did You did Yeah Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode. Available on Plus and Pro Plans. [50:00] puts a growing business on pause. Those are the days no one sees coming. But for more than 75 years, Cincinnati Insurance has been ready when they do. With deep expertise, coverage for businesses, homes, valuables, and more, Cincinnati Insurance works with independent agents who take the time to understand what matters to you. Together, they help protect more than what you own. They help protect what you're building toward. And when a bad day does come, Cincinnati Insurance has real people there to help make things right. Because planning for the future isn't only about knowing what's next, it's about making sure you're ready for what you can't predict. Let Cincinnati insurance make your bad day better. Find an independent agent at c-i-n-f-i-n.com. This is Alexis Christoffers for Bloomberg Surveillance. AI is everywhere. Outcomes are not. Not because AI doesn't work, but because AI hasn't reached the workflows yet. The campaigns, the launches, the quarterly planning. Those workflows are still run by humans alone. AI is making individuals faster, but it's not making businesses more productive. That's the gap Asana is built to close. Asana is the operating system for human agent teams your easy button for AI productivity across every team Ready to go AI teammates pre for marketing ops and IT No prompt engineering no setup They show up where the work is happening, already onboarded in your workflows, ready to deliver. And the more they work with your team, the smarter they get. With Asana, your whole company can work on the same plan towards the same goal, whether you're a team of 10 or 10,000. Asana, where humans and agents workflow together. Try it at asana.com. That's A-S-A-N-A dot com. Since we're talking about some of these questions and measurement and limited data, I wanted to get your take on, I think you may have, I don't know if you saw it, there are these questions about, it came up with NVIDIA specifically, and whether some of the chips that they design, but which are manufactured elsewhere, largely Taiwan, etc., whether they're being adequately captured in data. And there was this report out that said, oh, we should have added 0.3% to GDP, which is not that much in the sense that, like, oh, the big techno-optimists were right. No, but it's not nothing. I'm curious what you think. It is real money. On the other hand, when I look at it, I say, well, the market certainly noticed that NVIDIA has been doing very well The Fed doesn target GDP so it doesn really imply anything about oh we would have done something different Tell us what your thoughts on this discussion So I mean GDP measuring how much income we get for how much production and how much labor that what GDP is meant to do on a national basis and it is much more science than art, but there are pieces of it where it's not very simple. And so there are multiple definitions. There's something called GDI, and then there's gross final demand, and there's these various things. And they're all meant to be a check on each other. And over time, they basically move together. So what you're saying, Joe, about this discussion of mismeasuring of, say, NVIDIA or the other top tech companies, again, doesn't really change the fundamental path. But going back to what you both raised, if we're trying to assess how much we've seen a big transformation in productivity, this matters. If this 0.3%, say a year, was overlooked and it is directly attributable to the AI sector then that tells a different story about productivity So it does matter A colleague of mine I already mentioned Martin Chorzempa at Peterson is doing some work on this Former Fed official at Peterson, Joe Gagnon, has also written about this. So there's, I mean, there are a lot of subtleties, which I don't even understand, but there are basically two issues. One is sometimes when companies that are based in the U.S., using U.S. technology, produce stuff abroad, the allocation of how much stuff is actually produced in the U.S. versus not, and therefore how, whether it's GDP versus GNP, what's imports, what's exports, gets messed up. And so with these very complicated supply chains, there is an issue of trying to figure out what stuff gets allocated to the domestic pile versus the import pile. This is the old value add argument. Yeah, sorry. Again, I'm worried I'm going to screw up on nuance. So look on our website for a paper by Joe or a paper Joe Gagnon or Martin Shenzhenko or a tweet. I'm afraid I don't want it. But just essentially, there is a discussion over how much is domestic versus national. But as you said, Joe, in the end, the market is not a matter of. [55:00] you're either selling stuff or you're not. I mean, NVIDIA's bottom line actually doesn't change based on any of this. Right. We all know they're making a ton of money regardless of how the government is. But also, the people who are working at NVIDIA designing the chips in the U.S. are still getting paid. Absolutely. And all the money that comes into NVIDIA, you know, some of it goes to the shareholders, some of it goes to investment, some of it goes to the workers, you know. So again, this is worth worrying about, but this is not a major distortion. The second thing, though, which is the more contentious part, is there are always in technological revolutions, there are recurrent statements that GDP or whatever economists are using are just not capturing the true value. And usually the economist's response to that is some combination of, A, well, we're capturing the true market value. So if people are getting more out of it than what they're paying in the markets, great, but the market value is what we're supposed to think about. So that's one response. The second response is oh well there are these various things like hedonic indexing where we try to adjust for quality And this shows up more on the inflation side than the growth side but that helps you figure out what real growth and what not So let look at this specific sector and try and fix that So there is still a controversy about that. I mean, but this goes back, Tracy, to what you were saying about the timing and the productivity or the J-curve, right? So we already have GPT, Claude, whatever. I don't want to cite any particular brand name. We have the AI. and you or I can go in and type get me the cheapest possible ticket to Jackson Hole and I don't want to have to think about it but don't seat me next to a baby and that provides utils Tracy when she says don't seat me next to Joe I thought it was don't seat me next to you we really need another four hours on a plane no no no no and so that's giving me utils you know it's saving me some time it gets me a better experience I don't get annoyed with the AA or United or Delta website because somebody else is dealing with it for me does that show up in GDP? no so this is why we usually think the big productivity gains and the disinflation stuff comes further down the pipeline It when pretty on the track I should say not the pipeline when the businesses start transforming So, like, it was really cool that Intel and Texas Instruments and whoever else was building chips in the 90s, you know, were following Moore's Law and all that, and that showed up a bit in GDP. But the gains really happened when McDonald's and Walmart and UPS and everybody transformed their businesses to take advantage of it. Joe, do you remember when Goldman Economist, I think it might even have been Hatshia, yeah, they put out a note saying that productivity was being mismeasured because the graphics in the new Grand Theft Auto game, I can't remember what edition it was at that time, were so much better than the previous graphics. We should have them back on now that the new one's coming out. Because the new GTA 6 is coming out. Yeah, we should do that. All right. I have just one more question. We could talk for hours, and unfortunately we don't have that much time, But just in terms of the pure inflation outlook, the last time we had you on, it was right after you published a paper saying that you thought inflation was going to be 4% by the end of the year. I can't remember if you were looking at PCE or CPI. No I was saying at that time I was saying 4 on PCE and a little higher on CPI Yeah okay Well I mean at least on CPI we got to over 4 earlier in the year and I think a lot of people were surprised by that direction of travel What's your thinking now? I guess, thank you for citing that, Tracy. I guess what I was saying, and it was things I was saying, and then a joint piece with Peter Orszag of Lazard, what we were saying was, going back to where we were with the criticisms of the Powell Fed that I think were legitimate. The U.S. labor market is much more resilient than they thought. Credit was much more available or in the terms, the financial conditions were not as tight as they thought. Fiscal policy was a little looser than we thought. And the Fed, having not brought down inflation for 64 months or whatever it was then, 55 months, did have some momentum built up. And then additionally, one thing we talked about was, as I said, I think there was a J curve in the impact of tariffs and anti-migration policies over time. It took time for it to kick in. But anyway, the big point was, whenever the next inflation shock comes, it's going to go worse. I had no idea that the President of the United States would bomb Iran and ignore it. [60:00] or the fact that disrupting straight-seal hormones might have some inflationary effects beyond whatever else you think of it. But I knew there was going to be an inflation shock, and we were primed to have more. So the next couple of months, inflation may trickle down a tenth or two because of the energy market, whatever. But as Chair Walsh said, as many FOMC members have said, this is not good. You've got persistence. You've got persistent inflation and essentially the core measures that are services and not imported goods. You've got, as the chair cited in a speech, you've got three, six, 12 months where the moving average looks like it's going up. It's in the high threes rather than four, but it's going up. So my view is the Fed is going to hike. If they don't hike in September, they're certainly going to hike in December. I expect if they hike in September, they're still going to hike in December. And so six months from now, Fed funds will be 75 pips or 100 pips higher than it is now. And inflation will start coming down for reals as the kids say but until then we going to be in this three and a half to four and a half range with some upside risk All right. Adam Posen, always great catching up with you, especially this beautiful location. Thank you so much for coming back on Lopop. Thank you for having me. I'm a big listener. Thank you so much. Thank you. Love to hear it. Tracy, I love chatting with Adam. It's so good. It's always really good. I'm really glad we made it happen. You know what's interesting? I'll be curious to see how Powell's legacy ages. I think there's going to be, as Kevin Wurst said, 64 or 65 months of above target inflation. When you think back at the last several Fed shares, a lot of generally like the one who people like really think that was a bad tenure was arthur burns right despite and bernanke has held out like people have think of him fondly despite the fact that he presided over a financial crisis and several years of high unemployment um it doesn seem like from a reputational perspective high unemployment under your watch gets sort of penalized the same degree as high inflation People really seem to hate inflation You right Like it kind of skewed Yeah it clearly, it clearly is. So, you know, between like the questions like, well, did the Fed gather this sense of momentum where people were not dissenting for maybe reasons that were sort of inevitable or structural, et cetera, and you go over five years of above target inflation? I just think it'll be really interesting to see how people are talking about Powell in the years ahead. It was also interesting to hear that there seems to be a sort of unspoken norm that you can only have a handful of dissents per meeting. The other thing I was thinking about, just on the messy jobs note, there is this thinking out there that, oh, well, for instance, contractors can get replaced because now you can just ask ChatGPT how you're supposed to, I don't know, fix your toilet or something like that. But I was thinking, so my husband and I, we were doing a project where we were roofing an outdoor shed, and we asked ChatGPT how to do it. And we watched a couple YouTube videos, and it told us to use this one product, like to stick on the shingles and things. And then after we had started doing it we realized that that product only worked in environments that were above 60 degrees Fahrenheit and we were in New England And the bots and the YouTube videos were incapable of spotting that we were in fact in Connecticut versus Florida. So, you know, there's some nuance left for humans. Yeah, I think that's right. I feel strongly like for all the extraordinary things that the models can do that there are just so many of these minor things that we don't even articulate about what human judgment looks like, that they're not there yet. They may yet get there. They may get there by the end of the year, but they're not quite there yet. But just on the conversation, I do think it was interesting. The term that I've been thinking about with the Warsh speech, I found it to be hawk-ish, which is like with a dash between the hawk and the ish, as in it was clearly more like, yes, clear that there is a plan to fight and that there is going to be work to do to get inflation because he said the two things that were key, which is inflation is going in the wrong direction and policy has been insufficiently restricted, which therefore the third thing that the silent part is therefore high grades, which he didn't say. So that is hawkish. On the other hand, it was not hawkish in this. [65:00] that we are now about to embark on an aggressive, it was not Apollo 2022, where it's just, we're going to get inflation down and I'll see you next year. Well, it also begs the question, okay, you could have said the exact same things at the July meeting, and yet he opted not to hike because he said he wanted to see more information, but more information is always coming out. And also in this environment, it feels like more shocks are also constantly happening. And so, you know, you have to wonder what the threshold actually is. Yeah. Well, anyway, I'm glad. Big year, much to think about. All right. Shall we leave it there? Let's leave it there. This has been another episode of the Odd Thoughts podcast. I'm Tracy Alloway. You can follow me at Tracy Alloway. And I'm Jill Weisenthal. You can follow me at The Stalwart. Follow our producers, Carmen Rodriguez at CarmenArmin, Dashiell Bennett at Dashbot, Kale Brooks at Kale Brooks, and Kevin Lozano at Kevin Lloyd Lozano. And for more Odd Thoughts content, you should check out our daily newsletter. You can find that at Bloomberg.com forward slash oddbots. And you can chat about all of these topics 24-7 in our Discord, discord.gg slash oddbots. And if you enjoyed this conversation, then please leave a comment or like the video or better yet, subscribe. Thanks for watching and listening. If you listen to financial news you know a lot of time is spent thinking about what next The next opportunity. The next investment. The next move. But sometimes what matters most is being ready for what you never saw coming. For more than 75 years, Cincinnati Insurance has worked with independent agents to help protect businesses, homes, valuables, and more. Because planning for the future isn't only about knowing what's next. It's about making sure you're ready for what you can't predict. Let Cincinnati Insurance make your bad day better. Find an independent agent at CINFIN.com. Coffee genius here. Most people see a busy cafe, but I see precision at every step. Thanks to Genius from Global Payments. Transactions? Instant. Inventory? Precise. Operations? In sync. Absolutely genius. From sold-out crowds worldwide to managing the morning rush, Genius keeps operations running smoothly. One portato. Flawless pour, perfectly timed. Just beautiful. Big League reliability for any business. That's Genius. Join Bloomberg for the Canadian Finance Conference, proudly sponsored by National Bank of Canada Capital Markets on September 29th in New York. Hear from influential corporate and government leaders as they discuss the strategies shaping Canada's economic future. Connect with senior decision makers, gain actionable insights, and be part of the conversations driving business forward. Register at bloomberglive.com slash Canadian Finance. That's bloomberglive.com slash Canadian Finance.