Pepperstone logo
Pepperstone logo
  • 中文版
  • English
  • 交易方式

    概览

    定价

    交易账户

    Pro

    高净值客户

    好友推荐计划

    活跃交易者计划

    交易时间

    维护计划

  • 交易平台

    概述

    交易平台

    集成

    交易工具

  • 市场与产品

    概述

    外汇

    股票

    交易所交易基金

    指数

    大宗商品

    货币指数

    指数差价合约股息

    股票差价合约股息

    差价合约远期

  • 市场分析

    概述

    市场导航

    每日简报

    会见分析师

  • 学习交易

    概述

    交易指南

    网络研讨会

  • 合作伙伴

  • 关于我们

  • 帮助和支持

  • 中文版
  • English
  • 开启 Webtrader

  • 交易方式

  • 交易平台

  • 市场与产品

  • 市场分析

  • 学习交易

  • 合作伙伴

  • 关于我们

  • 帮助和支持

分析

FOMC

Why Real Rates Should Be The Focus

Michael Brown
Michael Brown
Senior Research Strategist
2024年2月6日
Share
As the FOMC seek to fire the starting gun on the easing cycle – albeit not as early as some had anticipated after Chair Powell’s recent pushback on the idea of cuts as soon as March – now seems an opportune moment to take stock of the overall monetary policy stance, and how said stance may evolve over the coming months.

Firstly, a rather crude look at the target range for the fed funds rate is useful. Though it tells us little about the relative policy stance, it does help to evidence the extent of the tightening cycle that the Fed wrapped up last July, with over 500bp worth of hikes – including the most in a calendar year (2022) in four decades – taking the fed funds rate to its highest level since 2001.

Preview

While interesting, from a historical standpoint, such a chart tells us little about the overall policy stance. To gleam such information, one must look at the ‘real’ fed funds rate; the nominal rate, above, adjusted for a suitable measure of inflation. This helps one to gauge whether policy can be perceived as ‘tight’ or ‘loose’, particularly when compared to the natural rate of interest – more commonly known as r*.

Again, rather crudely, a simple real fed funds rate can be constructed by subtracting the YoY change in CPI – the most widely used inflation metric – from the fed funds rate. Doing so, per the below chart, one can see that policy is at its tightest since the middle of 2009.

Preview

However, as alluded to above, there are many ways to calculate such a real rate, and many ‘suitable’ inflation measures that one may choose to use in such a calculation.

Perhaps the most important of these, though, is the one that Chair Powell himself has flagged as his preferred such indicator. Last May, in remarks that seemingly flew under the radar at the time, Powell noted that his preferred definition of the real fed funds rate is to subtract from it the one-year inflation breakeven. This being a market-based measure of expected inflation, derived from the difference in yield between nominal and real interest rates on Treasuries over said tenor.

Preview

As is clear to see, using this metric produces a significantly different chart. Using Powell’s preferred measure, one can see that policy actually achieved a much ‘tighter’ stance during the second half of 2023, before having loosened rather notably of late, as inflation expectations re-anchor towards the Fed’s 2% target, with disinflationary momentum continuing. An aside, it’s also interesting how Powell’s measure points to policy having tightened more substantially during the 2015-18 hiking cycle, and subsequent market tantrum over Powell’s ‘autopilot’ balance sheet comments.

All of this may sound rather academic, and akin to something you’d read in a macroeconomics textbook, however there are important market, and policy, implications stemming from these various measures, and where they currently stand.

To digest these, one must recall the aforementioned components which comprise a real rate – namely, the nominal fed funds rate (set 8 times per year by the FOMC), and an appropriate inflation measure. The latter factor is the focus here, for now, as, were the FOMC to hold the target range for the fed funds rate steady, as inflation continues to return to the 2% target, real rates would rise, making the overall policy stance tighter and more restrictive, right at the latter stages of the economic cycle, as the Fed are attempting to stick the ‘soft landing’.

The below attempts to illustrate that point in graphical form, using the crude real fed funds rate per CPI discussed earlier.

Preview

This is where things get interesting.

If one were to set policy using the CPI-based real fed funds rate (or PCE-based, if the Fed’s favourite inflation gauge is preferred), the clamour for rate cuts would likely grow to a deafening level in relatively short order, as the continued fall in headline inflation – swaps see 2% by August – pushes policy into an increasingly restrictive setting.

However, if one were to instead set policy using Powell’s preferred model, based upon the 1-year breakeven rate, the story would likely be very different. As the earlier chart shows, per this measure, the real fed funds rate has actually been falling since the fourth quarter of last year, with policy having therefore – in contrast to many perceptions – become looser over the same time period.

The implications of this are important. As we heard from the FOMC last week, a firmly ‘data-dependent’ stance has been adopted, with the Committee seeking additional inflation data to provide ‘confidence’ that a return ‘towards’ 2% is on the cards. Unsurprisingly, there was little clarity around how much data, or how many more ‘good’ inflation prints would be required, before such confidence was achieved.

Despite that, the market-based real fed funds rate implies that policy is not in fact becoming more restrictive as price pressures fade. This means that, using such a model, policymakers are likely to be able to wait longer than they otherwise might to gain such confidence, threatening the OIS curve’s current pricing of the first 25bp cut being delivered at the May FOMC meeting, with a further four cuts of the same magnitude to then follow before the end of the year.

Preview

The pricing out of said cuts, were it to occur, is likely to provide a further fillip for the greenback, which already sits comfortably in positive territory on a YTD basis against all G10 peers, while also posing a downside risk for Treasuries, particularly at the front-end and in the belly of the curve.

Whether equities are able to whether such a potential hawkish repricing as well as they have done so far this year remains to be seen, though likely requires continued strong performance from the ‘magnificent seven’ stocks (or six, ex-TSLA) in order to do so.

Preview

There is a final important point to make on cuts. As the many charts above should prove relatively easily, the Fed cutting rates as the year progresses does not equate to policy moving to an outright loose stance.

Instead, the cuts that the FOMC – and most other G10 central banks – are set to deliver represent a move back towards a more neutral policy setting, dialling back the necessary, and rapidly-introduced restriction that stamped down on post-covid inflation now that price pressures have abated.

Such context is important, as one should not view the start of this rate cut cycle as a sign that we are soon to return to the post-GFC days of zero, or negative, interest rates – ZIRP and NIRP to coin the vernacular. Instead, policymakers are more likely to seek to return rates to r* (a fuzzy concept that refers to a level of the real fed funds rate which would keep activity at potential and inflation low and stable), and then keep them there. Though difficult to measure, r* in the US is currently estimated by most models to be around 1-2%, implying a 3-4% fed funds rate, if inflation were to return precisely back to target.

This means that, for the first time in many years, capital actually has a cost once more, and cash is likely to offer a more attractive option than over the last decade or so. Hence, structurally higher bond and equity volatility is likely to prevail over the medium-run, as markets adjust to this new, but oh so old-fashioned, dynamic.

Observing the Above

While the charts in this article have been created using a Bloomberg Terminal, it is possible to obtain some of the same information using TradingView.

This video covers the topic in substantially more detail - https://www.tradingview.com/chart/US05Y/uUoUKxcw-Rates-Obsession-a-pro-interest-rates-set-up-on-TradingView/


Related articles

Dollar’s Bull Run Shows Few Signs Of Slowing

Dollar’s Bull Run Shows Few Signs Of Slowing

USD
Trader Thoughts - 5 charts & ideas front of mind

Trader Thoughts - 5 charts & ideas front of mind

Charts
Examining Where The Path Of Least Resistance Leads

Examining Where The Path Of Least Resistance Leads

Equities
USD

此处提供的材料并未按照旨在促进投资研究独立性的法律要求进行准备,因此被视为营销沟通。虽然它并不受到在投资研究传播之前进行交易的任何禁令,但我们不会在向客户提供信息之前谋求任何优势。

Pepperstone并不保证此处提供的材料准确、及时或完整,因此不应依赖于此。无论是来自第三方还是其他来源的信息,都不应被视为建议;或者购买或出售的要约;或是购买或出售任何证券、金融产品或工具的征求;或是参与任何特定交易策略。它并未考虑读者的财务状况或投资目标。我们建议此内容的读者寻求自己的建议。未经Pepperstone批准,不得复制或重新分发此信息。

其他网站.

  • The Trade Off
  • 合作伙伴
  • 组.
  • 职业生涯

交易方式

  • 定价
  • 交易账户
  • Pro
  • 高净值客户
  • 活跃交易者计划
  • 朋友推荐
  • 交易时间

平台

  • 交易平台
  • 交易工具

市场与符号

  • 外汇
  • 股票
  • 交易所交易基金
  • 指数
  • 大宗商品
  • 货币指数
  • 加密货币
  • 差价合约远期

分析

  • 市场导航
  • 每日简报
  • Pepperstone 激石脉搏
  • 会见分析师

学习交易

  • 交易指南
  • 视频
  • 在线讲座
Pepperstone logo
support@pepperstone.com
1300 033 375
Level 16, Tower One, 727 Collins Street
墨尔本, VIC 澳大利亚 3008
  • 法律文件
  • 隐私政策
  • 网站条款与条件
  • Cookie政策
  • 举报人政策

风险警告:差价合约(CFD)是复杂的工具,由于杠杆作用,存在快速亏损的高风险。 81.3% 的散户投资者在于该提供商进行差价合约交易时账户亏损。您应该考虑自己是否了解差价合约的工作原理,以及是否有承受资金损失的高风险的能力

风险警告:差价合约和外汇交易是有风险的。它不适合每个人,如果你是一个专业客户,你的损失可能大大超过你的初始投资。你并不拥有相关资产或对其拥有权利。过去的业绩并不代表未来的业绩,而且税法可能会改变。本网站上的信息是一般性的,没有考虑到你的个人目标、财务状况或需求。你应该通过审查我们的目标市场的确定文件来考虑你是否属于我们的目标市场,并阅读我们的PDS和其他法律文件,以确保你在做出任何交易决定之前充分了解风险。我们鼓励你在必要时寻求独立建议。

Pepperstone Group Limited位于澳大利亚维多利亚州墨尔本柯林斯街727号第一座16楼,邮编VIC 3008,并由澳大利亚证券和投资委员会(Australian Securities and Investments Commission)许可和监管。 本网站上的信息以及所提供的产品和服务均不得分发给任何国家或地区(如果其分发或使用违反当地法律或法规)的任何人。

© 2025 Pepperstone Group Limited | 澳大利亚公司注册号 (ACN) 147 055 703 | 澳大利亚金融服务牌照号(AFSL) 414530