Author: Cailian Press
At 10:00 PM Beijing Time on Friday, Federal Reserve Chairman Kevin Warsh appeared at the Jackson Hole Annual Meeting, delivering a speech titled "In Our Time".
Overall, Warsh's Jackson Hole speech delivered a relatively clear and cautiously hawkish signal. He believes that the U.S. economy and labor market remain resilient, and the current financial environment can hardly be described as significantly restrictive, while inflation remains notably above the Federal Reserve's 2% target. Therefore, the issue of prices should continue to be the primary focus of monetary policy.
The Federal Reserve Chairman emphasized in his speech: "My criterion is: We must have confidence that underlying inflation is clearly and moving towards our target fast enough. Otherwise, we have more work to do."
Warsh also stated that despite the summer CPI and PCE price data being better than expected, "they have not convinced me that a meaningful improvement in the underlying inflation trend has occurred".
Regarding external criticism of his "persistent refusal to give forward guidance", Warsh took this opportunity to provide an unprecedented and in-depth explanation.
Warsh believes that forward guidance is necessary during crisis periods but should be significantly weakened during normal times. He argues that hinting or even approximating a commitment to future interest rate paths prematurely, while ostensibly enhancing transparency, may actually create new misunderstandings: on one hand, it constrains the Federal Reserve's future space for flexible decision-making based on economic changes; on the other hand, it also causes markets to excessively trade around "guessing the Fed" rather than independently judging economic fundamentals.
He is particularly wary of the resulting "hall-of-mirrors problem"——markets price based on the Fed's guidance, and the Fed in turn references market prices for judgment, potentially causing both sides to overlook new economic changes simultaneously.
Therefore, Warsh neither supports normalized forward guidance nor is willing to provide a mechanical policy "reaction function". In contrast, he prefers reducing pre-commitments, allowing markets to form their own judgments, while the Federal Reserve, based on real-time data, trends, and more robust policy rules, maintains sufficient freedom at each moment when a decision is truly needed.
As of 10:45 PM Beijing Time, after Warsh's speech, the CME "FedWatch Tool" indicated that the probability of a Fed rate hike in September has warmed to nearly 60%, compared to only 35% yesterday. Spot gold plunged by $50 in the short term, with the latest quote falling to around $4,550 per ounce.
(Source: TradingView)
Below is the full translation of Warsh's speech (The speech transcript source is from the Federal Reserve official website, assisted by AI translation)
"In Our Time"
Federal Reserve Chairman Kevin Warsh
August 28, 2026
Remarks at the Economic Policy Symposium "Financial Innovation: Implications for Payments and Policy" in Jackson Hole, Wyoming. The symposium is hosted by the Federal Reserve Bank of Kansas City.
Thank you. It's great to be back, and great to see so many familiar faces. I've been looking forward to this weekend — where else could be more fitting to mark my 100th day as Chair of the Federal Reserve?
For the gracious hospitality here, each of us owes thanks to Kansas City Fed President Jeff Schmid and his colleagues. Jeff, a sincere thank you to all of you.
Jeff and the other conference organizers have also arranged some recreational activities for later today. I suggest everyone be very careful in their choices.
As I learned years ago, there are two very different kinds of hikes you can take on the trails around Jackson Hole. I can sum up my past hikes with former Fed Vice Chair Don Kohn in two words: I survived. Those marathons of willpower—'death marches'—revealed a side of Don I was utterly unprepared for.
Then there's the other kind of hike—I'd associate that with my old colleague, former Fed Chairman Ben Bernanke. With Ben, the pace is far more leisurely, just a casual stroll along the winding paths of the Rockefeller Preserve.
So, before you set out, check your fitness level, and ask yourself: 'Is it a Kohn day or a Bernanke day?'
One of the best things about this gathering is that it helps us all clear our minds and think more clearly about the world and the era we inhabit. For me, this is the right place, and you are the right audience, to delve into the most important ideas.
"Innovation" is the theme of this conference. I believe that the public and the market, with their collective wisdom, have recognized that innovation in the way the Fed implements policy will help us achieve both maximum employment and price stability.
Let me briefly outline what I will cover this morning. You can call it an outline... or a hiking map... but please don't call it "forward guidance."
First, I'll discuss several long-term issues the Fed is examining, including the latest general-purpose technology—Artificial Intelligence (AI)—and where it might take the economy.
Then, I'll talk about the practice of forward guidance and the interaction between central banks and financial markets.
Next, I'll introduce some core principles that I believe should guide the implementation of monetary policy.
Finally, I'll discuss my assessment of the current economic situation.
Preparing for the Future Policy Environment
Against the backdrop of the ever-constant Teton Mountains, we gather here to examine an economic landscape that is anything but static.
Not so long ago—just before the 2008 crisis and for the subsequent decade—economists and policymakers were discussing "secular stagnation" and "global saving glut." A widely accepted view then was that there simply would not be enough attractive investment opportunities, and excess capital would sit on the sidelines for a long, long time. All the good things had already been invented. Therefore, economic growth would be sluggish and slow.
However, times have indeed changed. We have reached a historical turning point.
Take the most obvious example: Artificial intelligence—a name with 80 years of history now used to refer to the latest technological wave—is advancing even faster than the most enthusiastic proponents predicted just a few years ago.
The potential for significantly higher economic growth is rising. An expanding pool of capital is pouring into all sorts of AI-related infrastructure. A kind of "super Moore's Law" seems to be at play. At the same time, scaling laws are changing the methods and speed of innovation.
Capital is combining with labor to create the large language models at the heart of AI. Users purchase tokens for access to these models. According to reports, the annualized token sales of just two leading AI labs have already exceeded $100 billion, an increase of more than 500% compared to a year ago.
The Federal Reserve is watching all of this closely. We recognize that AI is a new variable—and potentially a new factor of production—that will have implications for the economy and the implementation of monetary policy. This also opens up several major lines of inquiry:
Will the application of AI drive a significant and sustained increase in productivity across the economy? If so, when?
Will the use of tokens primarily complement labor or compete with it? Will the next generation of AI models require even higher capital intensity, or will the models themselves ultimately help design solutions that are less capital-intensive?
Other unresolved questions include what market structure will ultimately emerge. It is unclear where the returns to capital will ultimately accrue, nor how long this process will take. In the early stages, how much of the economic surplus will flow to the owners of scarce assets—AI labs, chip manufacturers, energy producers, and cloud providers? Over time, how much value will ultimately accrue to businesses and consumers? What do these changes mean for workers? And what are the broader implications for the employment goal within the Fed's mandate?
Similarly, we don't yet know what the equilibrium price for tokens will be. Will different types and qualities of tokens emerge in the future, such that people are willing to pay increasingly higher prices for access to the most cutting-edge, best models? Will the price of tokens for older-generation models eventually fall to their marginal cost?
We will delve into these questions with the help of a "Productivity and Employment Working Group." I recently had initial conversations with the leads of this group and four other working groups, and their progress is encouraging.
To be clear, however, recommendations from these groups will be submitted in the future and will not influence decisions we make in the current policy environment. But I believe that investing in such thinking today, for tomorrow's policy challenges, will leave us much better prepared.
Forward Guidance and Alternatives
While these working groups do their work, I have not been idle. I have already begun to foster innovation at the Fed, to truly adapt the institution to its mandate. To give one example, I have taken steps to alter the form and function of what's known as the Federal Reserve Chair's "forward guidance." As you may know, I have long felt uneasy about announcing future policy decisions prematurely. I lean towards a different path... and I'll explain why.
Transparent communication about future policy decisions is not a virtue in itself. Communication must serve the Fed's most important responsibility: getting monetary policy right.
During the global financial crisis, my colleagues and I at the time established forward guidance as a regular practice. At that time, it was essential, and we rolled it out with fanfare. But like other legacies from past crises, I believe this practice has outlived its usefulness.
In normal times, the role of forward guidance should be limited and clearly bounded. Otherwise, it can create confusion in the name of seeking clarity. Excessive disclosure of policy discussions and making too many commitments about future policy decisions can mislead markets, businesses, and households. And I believe that when policymakers make near-commitments about interest rates across the business cycle, we actually limit our freedom to make the right choice at the moment a decision is truly needed.
To get policy right, we must also properly manage the relationship between financial markets and the central bank. The Fed needs clear market signals, and these signals should be as unfiltered as possible... including the internal structure of markets... the level and movement of asset prices across sectors... the price and trading volume of U.S. Treasuries... the foreign exchange value of the dollar... the cost and availability of credit... and the prices of a broad range of commodities.
These and other indicators should help the Fed assess near-term economic activity and inflation prospects throughout the business cycle. They should also reveal the state of broader financial conditions... and the risks and uncertainties within the financial cycle.
At the same time, market participants themselves should be tracking real information across the economy. They should form their own judgments; form their own expectations about output, employment, and inflation; and always pay close attention to risk.
The Fed should be humble but never naive. The Fed plays a crucial role in the economy and markets, and our policy tools are powerful. We determine the path of short-term interest rates. Therefore, market participants will always try to predict our next move. But we should not foster a mechanism where market participants primarily decide their next trade by second-guessing the Fed.
The economics literature has long described this distorting effect: it's known as the "hall-of-mirrors problem." If markets rely heavily on Fed guidance, and the Fed in turn relies on market prices, then we are all more likely to be blind to new developments... more likely to be caught off guard when conditions suddenly shift... and more likely to make policy mistakes.
Ironically, market participants may not bear the greatest cost of the "hall-of-mirrors problem." Those most severely harmed are likely those without financial assets. If the Fed misjudges inflation and misjudges the economy, who suffers the most? Not the high-net-worth individuals in financial markets. The ones who ultimately have to face excessive inflation or suddenly unstable employment are the hard-working ordinary Americans.
So, if forward guidance is not suitable for normal times, shouldn't the new Fed Chair at least promise a clear reaction function? Certainly, he should tell us where rates would go if the data came in clearly hot or cold.
I wish our understanding of the economy were precise enough to offer a mechanical, fail-safe answer—like a simple rule akin to the Taylor rule. But our knowledge is far from that point—at least not yet—and the most important factors determining appropriate monetary policy themselves change over time.
Demonstrating the Fed's reaction function through predictions works better in theory than in practice, better in the lab than in real-world operations. I'm not the only one who has noticed this. For example, the forward guidance of 2021 likely slowed the Fed's subsequent policy response to high inflation.
During my tenure as Chair, my colleagues and I will strive to build more reliable models and more robust rules to guide policy decisions. We will undertake this work on the basis of the recognition that accurate economic forecasting remains a goal. As geopolitics, global supply chains, and technology change at such a rapid pace, humility about what we can and cannot know is wise.
In the same spirit, we should fully listen to a diversity of views on any issue that might influence Federal Reserve monetary policy decisions. If our goal is optimal decision-making, we should not exclude differing views on the economy.
So, what's a better path for policy? In the rest of my remarks, I'll share some core principles that guide my thinking on the proper conduct of monetary policy... and then fulfill my promise to discuss my assessment of the economy.
Core Principles
On to principles...
First, I note that in our business, people often mistake yesterday's news for what is happening right now. The real challenge is distinguishing between the two. In other words, we must constantly test reality to ensure we are not making future-facing policy based on data that is already outdated or inaccurate. Nor should we rely on isolated data points. Trends are what matter most. The Fed is a decision-making institution. We must choose amid uncertainty, so the data we rely on must be as relevant, timely, accurate, and directly usable for decision-making as possible.
Second, the Fed acts to ensure that aggregate demand in the economy roughly aligns with aggregate supply. However, what we can directly observe is only economic activity itself. We can never directly see what is truly happening on the supply side; we can only infer. Therefore, the assessment of the balance between aggregate supply and aggregate demand, both current and future, is inherently imprecise.
Third, this must not be misunderstood in any way: the Fed's 2% price stability target, as measured by the personal consumption expenditures (PCE) price index, is a firm, fixed target. Another aspect of this target must also be made equally clear: price stability is not automatic, and inflation is not inherently mean-reverting. Achieving price stability is the Fed's job.
Fourth, the Fed is equally responsible for achieving maximum employment. Achieving the two goals of the dual mandate over the medium term is not an "either-or" proposition. I do not see the Fed's dual mandate as conflicting. After all, high inflation itself severely undermines economic prosperity.
Fifth, short-term interest rates are the primary policy tool for achieving the dual mandate. Unconventional policies aimed at stimulating economic activity may be appropriate for true crisis periods but should be used sparingly, if at all, in other circumstances.
Sixth, money matters. This view is not currently fashionable, but my view is that there is certainly an important relationship between money and monetary policy. We should pay attention to money created by the central bank, and also to money from the banking and financial system. Indeed, financial innovation and other factors change the mechanism connecting the monetary base, the velocity of money, and the broader economy. But this is no reason to ignore that money ultimately has an impact on financial conditions and prices.
Finally, a quieter, more purposeful Fed in communication will be better able to achieve its goals. And whether we have fulfilled our responsibilities can be held accountable—this is the only true standard for testing our credibility. To borrow a phrase from General Chuck Yeager: "When the moment of truth arrives, there are either explanations or results."
The Current Economic Situation
So, under these principles, how do I assess the economy today? What is really happening out the window?
As you may have seen from the July meeting minutes, the unanimous judgment of the FOMC is: a stable labor market, robust economic output, but inflation remains too high. The majority of my colleagues and I believe the wiser course is to wait for more new information between meetings—especially considering potential new developments in supply chains, investment flows, and geopolitics—before judging whether adjusting interest rate policy is appropriate. At the same time, we have also jointly indicated that we stand ready to act as needed based on the evolving situation.
Personally, I am impressed by the overall performance of the U.S. economy today, and it seems to have strengthened. One standard for judging a strong economy is how well it withstands shocks. From that perspective, both Main Street in the real economy and Wall Street in the financial markets have shown remarkable resilience.
A few observations:
Business capital expenditures—the "seed corn" for future economic growth—are growing rapidly. The four-quarter increase in equipment and intangible assets investment is about 9%, the highest growth rate since 2021. More than half of this year's capital expenditure growth can likely be attributed to AI-related infrastructure construction.
For S&P 500 component companies, profit growth over the past year exceeds 20%. Compared to historical levels, corporate profit margins are quite high. Overall stock market volatility is low. We are closely watching market internals, observing performance across sectors.
Market expectations for future growth in capital expenditures and corporate earnings are quite high. I will continue to observe the changes in their growth rates themselves, the so-called "second derivative." The follow-on effects—including on asset prices, business confidence, consumer income, and consumer spending—are also important.
Credit spreads for both corporate bonds and leveraged loans are near the low end of their historical ranges, and issuance volume in these markets this year has been quite strong. Stepping outside fixed income and looking at banking, the July Senior Loan Officer Opinion Survey shows banks telling us that credit standards for commercial and industrial loans are currently at the relatively easy end of the historical spectrum. This also helps explain why such lending has grown this year. Credit and lending markets show little sign of being constrained by monetary policy.
Certain sectors—such as housing and agriculture—are indeed under pressure. But overall, I would find it difficult to characterize broad financial conditions as "restrictive."
Despite various shocks, real consumer spending remains healthy, growing over 2% in the past four quarters. Combining consumption with the strong investment we observe, private domestic final purchases (PDFP) have also grown. Year-to-date, PDFP growth is close to 3%. Compared to Gross Domestic Product, this indicator typically contains stronger economic signals, and the trend it currently shows is also positive.
On the employment side of the Fed's dual mandate, the U.S. is currently performing well. The labor market is quite stable. The unemployment rate is currently 4.1%, still low by historical standards and has shown little significant change over the past few years. The four-week moving average of initial jobless claims—a time-tested, fairly robust real-time indicator—is currently near multi-decade lows.
In my view, the relatively low labor market turnover rate currently is partly due to a large-scale re-matching between employers and employees following the pandemic.
When labor supply is barely growing, monthly job gains will naturally be lower. There are always areas of the labor market worth watching—such as recently graduated young people. But overall, those who want to work are largely able to keep their jobs or find work. They may certainly worry about potential future disruptions in the labor market, but so far, I consider the U.S. labor market to be consistent with maximum employment.
But on the price stability side of our dual mandate, the data is more concerning. The Fed's preferred inflation gauge—the increase in the PCE price index over the past 12 months—is currently 3.7%, and the increase over the past six months annualized is 4.1%. Comparable measures for the Consumer Price Index (CPI) are also high, and both the core PCE and CPI inflation measures are elevated. None of these indicators is perfect, but they all tell a similar story: inflation remains above our 2% target. Therefore, the Fed's current primary focus should be on prices.
The policymaker's task is to identify underlying trend inflation, that is, the broad-based price changes across the economy after excluding various special, idiosyncratic factors. We need to judge whether underlying inflation is rising, falling, or stagnant. We want to understand not only the direction of change but also its speed. Each of the broad inflation measures mentioned has declined significantly compared to the 2022 peaks. But the progress made over the past two years has been quite limited.
Moreover, although the summer PCE and CPI data were better than expected, these data have not convinced me that a meaningful improvement in the underlying inflation trend has occurred.
The data show that wage growth is currently also moderate. But in tracking underlying inflation, wage growth has long not proven to be a reliable predictor of future inflation.
To assess underlying inflation, I find it very helpful to break down the 199 individual components of the PCE price index. Over the past 12 months, 54% of the items in the PCE basket had price increases exceeding 3%. This proportion is significantly lower than the post-pandemic peak of about 77%, but still well above the 20-year pre-pandemic average of 32%.
Looking at just the last six months, the conclusion is similar: 49% of the items in the PCE basket had annualized price increases exceeding 3%. Again, this is significantly lower than the post-pandemic peak but still at a fairly high level.
The recent rise in overall commodity prices is also noteworthy. We need to judge whether these current trends suggest upside risks to inflation.
Furthermore, whether the persistently elevated inflation data over the past five-plus years has seeped into people's expectations is also very important. The good news is that medium-term inflation expectation indicators overall remain stable. Inflation compensation measures in the swaps market also convey similar and strong signals.
Particularly in light of recent developments, market prices still reflect a confidence—a belief that we can achieve price stability. This reflects both the credibility of the Fed as an institution and aligns with the finest traditions of the Fed. And I can assure you... the market is right.
From the perspective of economic history, market-based inflation expectation measures have a characteristic: before they lose stability, they often appear very resilient, very solid. These expectations do not change easily, and currently they remain well anchored. But we must watch closely. Ensuring that inflation expectations do not become unanchored is the Fed's job.
There is one signal no one should miss: the sustained, elevated inflation over the past 65 months clearly lies at the feet of the central bank. And that is exactly where the responsibility should lie.
My criterion is: We must have confidence that underlying inflation is clearly and moving towards our target fast enough. Otherwise, we have more work to do. This is our work... our mission... and the responsibility we must fulfill.
Conclusion
Standing here today, I commit to a discipline, not to a specific policy decision.
In such a consequential era, my colleagues at the Fed and I are certainly not the first to hold these positions. We are determined to cherish the present and perform our duties to the highest standards we can achieve.
We approach our responsibilities with humility and firm resolve. Much depends on the choices we make. Sound monetary policy can help families and businesses thrive. When monetary policy is effectively implemented, it can expand and deepen the sources of U.S. economic growth... while helping to cement America's leadership in the world. And I know our country needs us to think carefully and act wisely.
It is a profound honor to serve the Federal Reserve again. For the encouragement and valuable counsel from my colleagues... and the support from many of you here today... I am deeply grateful. Thank you, and thank you for your patience this morning.
