Yes! I've been thinking a lot about your posts a few years ago pointing our that higher rates eventually mean higher long-run inflation and every time I see the inflation numbers today and hear the discussion I think exactly "we're in the second half of John's graphs!". Relieved to hear you see it the same. ... now the harder issue: What would YOU do if you were Fed chair (and FOMC would follow you 100%)?. Personally, I'm not sure since the Fed can't control the fiscal side, engage in micro deregulation, or end war with Iran. I hate to say it but holding steady for now seems like the best move, repeating publicly that the long-run target is 2%, and then praying CPI level drops post Iran and they can cut incrementally as inflation slows... (these hold constant that we are in Ample reserve framework with IOR)... thoughts?
Thanks John. you are the first to point out this short and long term relationship. But the analysis leaves out a Very important input in today world, unappreciated by economists of yore. the economy is a financial based economy…read debt. raising rates raises the price of everything one way or another in both the short and long term. so in the 6-24 month time frame you get a diminished investment , lower supply, higher input costs ie debt costs go up. Of course the fiscal problems, left as is, is truly inflationary and is greatly magnified by rising rates. the proverbial catch 22.
The answer is do nothing. The bottom line let the markets stabilize or reach a “equilibriun corridor” my term, . probably some bad stuff along the way but hopefully spread over several years. BTW this means making sure the market understands the is NO More Fed PUT. the market will then know the fed means business.
V interesting read. Agree that without fiscal reforms, maybe there’s a limit to what the Fed can do. However, the real conundrum to many is how one transitions from hawkish to “dovish-ish” as inflation remains well above target.
But Wash’s Fed couldn’t control rational expectation. Now bond vigilante throw the 10Y30Y bond because the Trump wants to melt debt by inflation, so I think to stop trump is to stop inflation.
and Wash’s Ai narrative isn’t good at inflation. Ai doesn’t show the tfp and b2c model. None b2c model can mean none people want this… by Jacob Schmookler. Thk.
The PPI, the PCE, and the CPI with and without Food and Energy all show inflation affected by tariffs (from April 2025 onward) and the Iran War. The Core CPI -- not including food and energy -- was coming down until recently and is now stuck around 2.5%. None of this has to do with the Fed; it is all due to Trump's policies. The tariffs are a tax, and Trump still hasn't given up on them -- and they are not a one-time event, they get transmitted through the wholesale (PPI) system as well as with consumers at the retail level (which also include businesses that consume). There is not much the Fed can do when you have inconsistency at the fiscal level.
“The Core CPI -- not including food and energy -- was coming down until recently and is now stuck around 2.5%. None of this has to do with the Fed; it is all due to Trump's policies.”
When you state opinion (Core CPI now stuck at 2.5% “all due to Trump policies”) as fact you both reveal a complete lack of epistemic humility and state a tautology.
We have had inconsistency at the “fiscal level” for at least 6 decades now; is your claim that during all that time there is not much the Fed could do?
There is a big difference between the inconsistency over the past decades and the fits that come out of this Administration on a weekly if not daily basis. That, the Fed is unable to moderate or counter; in my humble epistemic tautological state.
The *fiscal* situation is not changing on a weekly or daily basis, even if you are directionally correct (but hyperbolic) on the regulatory frequency change.
But that is moving the goalposts if the claim is “fiscal”.
And the economic/“fiscal” changes on this “regular” basis are quite small in the context of the U.S. economy. At most you can cite the “Liberation Day” and perhaps 2 or 3 other days during Trump 47 has having anything other than minor fiscal impact.
1. The effect of tariffs is not a one-time event. We have many economic players in the wholesale system that deal with contracts every time a transaction is processed. They can choose to absorb the extra cost or pass it along. And that decision is made by many activities in the logistics train.
2. Last week the news was the U.S. Treasury bailing out the Japanese Yen. The role of the Fed was to act as an agent for the U.S. Govt; the Fed had no role in the decision.
3. Week after week after week, the crude oil prices gyrate in a $30 range depending on Trump's state-of-mind.
4. States are constantly dealing with the inadequacies of federal funding during emergencies -- fires, floods.
Response of my AI Assistant (Gemini): "Yes. A coalition of 25 states and Washington, D.C., filed a federal lawsuit against the Trump administration, accusing the Department of Homeland Security (DHS) and the Federal Emergency Management Agency (FEMA) of unlawfully withholding millions of dollars in congressionally appropriated disaster-preparedness and emergency grants. The administration tied the release of these funds—which states rely on for wildfire management, flood and earthquake risk mitigation, and first-responder support—to compliance with unrelated federal policy demands. Specifically, the administration attempted to withhold at least 20% of the grants unless states agreed to enforce specific election rules (such as voter roll citizenship checks and transitioning to hand-marked paper ballots) and cooperate with federal immigration enforcement."
You can try doing some of your own research Andy G. All you have to do is read a newspaper or keep up with the news on a daily basis.
You write "But it’s harder to base an economic theory on the idea that people never ever catch on, never ever start thinking about future inflation rather than past inflation when they decide how much to pay for a mortgage, how much to borrow for a business loan, how much to save vs. go out to dinner. So, when they eventually catch on and start noticing the gulf between inflation and interest rates widening, as in my first graph, " so where does missing your target for five years running fit in? Why does anyone pretend that the Fed has credibility...OK a New Chair...Like Star Wars...A New Hope.. Is it enough?
Thank you for this clear analysis. Since for the past 4 decades, Fed-manipulated interest rates have been held to be "levers" in a policy machine, Warsh is apparently playing the role of Hippocrates, "first do no harm." Aside from some major shock, is there ever a reason for the Fed to move the policy rate? If one were a banker, jacking the IOER would be rewarding; would it encourage more reserves? Would that contract M1? M2?
That inflation was so quiet during the near decade of interest rates stuck at zero does sort of beg the question whether the Fed is doing any good by trying to move inflation around.
If deterrence only works when people believe the stick is real, then a “wisely cautious” hike is the one move that can’t actually deter anything — it just tells everyone the stick was never coming
Much of the news commentary about the mood of the public regarding Trump's economic performance seems to relate to consumers' memory of last year's prices on repeated purchases. This suggests that "inflation" may not actually be an economically objective phenomenon (although we have found many different ways "to measure" it). People seem to desire a Zero rate of change in annual average prices, particularly in food and fuel. Maybe the fear of "the zero boundary" is misplaced and the sacred 2% should be reconsidered. There is something unlovely about knowing the value of a Dollar will be halved in 36 years. Thank you for reminding us about 1947 prices.
Sure, fiscal reforms are important and can help stabilize inflation. However, the question is what monetary policy can achieve on its own. Accordingly, I assume that the IRFs in the top figure are plotted under the assumption that primary surpluses remain constant.
To prevent inflation from rising in the future following a monetary contraction, doesn't the model imply that the policy rate would need to decline much more rapidly than suggested by the figure and, in particular, fall below its initial level, converging back to the steady state in an oscillatory manner?
What would such a policy look like in practice? Would it imply that the Fed should begin lowering interest rates *before* it has achieved its inflation objective?
This is a simple policy, not an "optimal" policy. In this model like any other, give the Fed an objective and there is an "optimal" response to shocks. Whether the real Fed can implement that is a good question.
I have a question rather than a comment. If the central bank can do little about inflation, why does the central bank talking “tough” have any impact on inflation expectations? In your framework, if they talk tough and show a willingness to hike, wouldn’t that raise inflation expectations?
I’m not an economist but simplistically why doesn’t the Fed just suck money out of the economy via the banking system so prices can’t rise overall because there’s no money to pay higher prices? This would have a more general effect than interest rates, which can be very damaging for some individuals. Another mad idea is for the Fed to be able to add a Rainy Day Tax on income. The money is taken (which slows the economy) and is held in reserves to be released during the next slowdown as a stimulus?
A major mistake in statistical models is called the error of the omitted relevant variable. It’s nice to show graphs comparing one cause and an effect but its purpose should be limited to deciding whether to include that variable in a larger model. My analysis agrees with the result that higher real fed rates are associated with higher inflation two years hence but my 40 year model includes most of the usual suspects as variables too. And the results are good:R2 around .6 (0.7 if you omit the chaotic past 12 months). Besides the ornery behavior wrt Fed funds there are other surprises too. Hopefully I will have a paper by the fall.
Thanks for the article, it is a really interesting and undercovered point.
I wanted to make two questions/notes:
- One of the key assumptions is the long-run neutrality of money. I know that this is one of the most unquestionable ideas within classic economics. However, I wonder whether that is really applicable under interest rate targets. When the Central Bank alters or "manipulates" the interest rate, it distorts or alters real rates as well. Suddenly some investment projects could stop being profitable, or viceversa depending on the discount rate. In summary, the decision to change interest rates very likely has an effect on capital allocation decisions. It is also a fairly accepted fact that the stock of capital in an economy is a non-stationary series, it does have unit-roots, and it is not necessarily an ergodic process. In a less mathematical sense, it means that capital is long-lasting, and capital allocation decisions last very long (e.g. building a stock of housing, or building a chemical plant with a life of 50 years...). Those capital allocation decisions will therefore have long-run effects on the productive side of the economy, on the real side of things. Is really long-run neutrality true in this sense? I agree with the idea that units of measure do not change economic facts, but here it seems that the units of measure can distort or influence long-run decisions that do affect production in the long-run.
-My second point is a question about interest rate targets, monetarism and productivity improvements. From Warsh's reasoning, it can be understood that he believes that higher productivity lowers inflation (he is thinking in terms of M*V=P*Y). If Y grows faster so can M, however, is this equivalent to lower rates as Warsh implicitly assumes? Can we make money supply growth equivalent to lower interest rate?
Also, there are other economists and policy makers suggesting that in the medium-term, the policy rate should converge towards neutral, and if productivity is higher and demand for investment is higher so will be the neutral rate, therefore the Fed risks running loose policy if they cut. Which of these conflicting views do you think is right? Is it possible to conciliate these two totally opposed views?
Thank you. Always a pleasure to read your thoughts.
Really enjoyed this, Professor Cochrane — the "big stick" point is something I hadn't considered before. I'd been focused on whether Warsh sees today's inflation as more transitory than the consensus, not on whether the credibility behind his approach is actually backed by anything.
Yes! I've been thinking a lot about your posts a few years ago pointing our that higher rates eventually mean higher long-run inflation and every time I see the inflation numbers today and hear the discussion I think exactly "we're in the second half of John's graphs!". Relieved to hear you see it the same. ... now the harder issue: What would YOU do if you were Fed chair (and FOMC would follow you 100%)?. Personally, I'm not sure since the Fed can't control the fiscal side, engage in micro deregulation, or end war with Iran. I hate to say it but holding steady for now seems like the best move, repeating publicly that the long-run target is 2%, and then praying CPI level drops post Iran and they can cut incrementally as inflation slows... (these hold constant that we are in Ample reserve framework with IOR)... thoughts?
Thanks John. you are the first to point out this short and long term relationship. But the analysis leaves out a Very important input in today world, unappreciated by economists of yore. the economy is a financial based economy…read debt. raising rates raises the price of everything one way or another in both the short and long term. so in the 6-24 month time frame you get a diminished investment , lower supply, higher input costs ie debt costs go up. Of course the fiscal problems, left as is, is truly inflationary and is greatly magnified by rising rates. the proverbial catch 22.
The answer is do nothing. The bottom line let the markets stabilize or reach a “equilibriun corridor” my term, . probably some bad stuff along the way but hopefully spread over several years. BTW this means making sure the market understands the is NO More Fed PUT. the market will then know the fed means business.
Would the “Taylor Rule” be useful?
I'd like to see Taylor Rule calculated interest rates vs. history up from 1970 until now.
V interesting read. Agree that without fiscal reforms, maybe there’s a limit to what the Fed can do. However, the real conundrum to many is how one transitions from hawkish to “dovish-ish” as inflation remains well above target.
But Wash’s Fed couldn’t control rational expectation. Now bond vigilante throw the 10Y30Y bond because the Trump wants to melt debt by inflation, so I think to stop trump is to stop inflation.
and Wash’s Ai narrative isn’t good at inflation. Ai doesn’t show the tfp and b2c model. None b2c model can mean none people want this… by Jacob Schmookler. Thk.
I am not an economist. Is the “Taylor Rule” too simplistic?
Yes it shows between inflation and unemployment rate mechanic. But it only uses short term economy system not food at long term.
The PPI, the PCE, and the CPI with and without Food and Energy all show inflation affected by tariffs (from April 2025 onward) and the Iran War. The Core CPI -- not including food and energy -- was coming down until recently and is now stuck around 2.5%. None of this has to do with the Fed; it is all due to Trump's policies. The tariffs are a tax, and Trump still hasn't given up on them -- and they are not a one-time event, they get transmitted through the wholesale (PPI) system as well as with consumers at the retail level (which also include businesses that consume). There is not much the Fed can do when you have inconsistency at the fiscal level.
“The Core CPI -- not including food and energy -- was coming down until recently and is now stuck around 2.5%. None of this has to do with the Fed; it is all due to Trump's policies.”
When you state opinion (Core CPI now stuck at 2.5% “all due to Trump policies”) as fact you both reveal a complete lack of epistemic humility and state a tautology.
We have had inconsistency at the “fiscal level” for at least 6 decades now; is your claim that during all that time there is not much the Fed could do?
There is a big difference between the inconsistency over the past decades and the fits that come out of this Administration on a weekly if not daily basis. That, the Fed is unable to moderate or counter; in my humble epistemic tautological state.
PPI/CPI/PC https://fred.stlouisfed.org/graph/?g=1XOvQ
Crude Oil https://fred.stlouisfed.org/graph/?g=1XOts
The *fiscal* situation is not changing on a weekly or daily basis, even if you are directionally correct (but hyperbolic) on the regulatory frequency change.
But that is moving the goalposts if the claim is “fiscal”.
And the economic/“fiscal” changes on this “regular” basis are quite small in the context of the U.S. economy. At most you can cite the “Liberation Day” and perhaps 2 or 3 other days during Trump 47 has having anything other than minor fiscal impact.
1. The effect of tariffs is not a one-time event. We have many economic players in the wholesale system that deal with contracts every time a transaction is processed. They can choose to absorb the extra cost or pass it along. And that decision is made by many activities in the logistics train.
2. Last week the news was the U.S. Treasury bailing out the Japanese Yen. The role of the Fed was to act as an agent for the U.S. Govt; the Fed had no role in the decision.
3. Week after week after week, the crude oil prices gyrate in a $30 range depending on Trump's state-of-mind.
4. States are constantly dealing with the inadequacies of federal funding during emergencies -- fires, floods.
Response of my AI Assistant (Gemini): "Yes. A coalition of 25 states and Washington, D.C., filed a federal lawsuit against the Trump administration, accusing the Department of Homeland Security (DHS) and the Federal Emergency Management Agency (FEMA) of unlawfully withholding millions of dollars in congressionally appropriated disaster-preparedness and emergency grants. The administration tied the release of these funds—which states rely on for wildfire management, flood and earthquake risk mitigation, and first-responder support—to compliance with unrelated federal policy demands. Specifically, the administration attempted to withhold at least 20% of the grants unless states agreed to enforce specific election rules (such as voter roll citizenship checks and transitioning to hand-marked paper ballots) and cooperate with federal immigration enforcement."
You can try doing some of your own research Andy G. All you have to do is read a newspaper or keep up with the news on a daily basis.
You write "But it’s harder to base an economic theory on the idea that people never ever catch on, never ever start thinking about future inflation rather than past inflation when they decide how much to pay for a mortgage, how much to borrow for a business loan, how much to save vs. go out to dinner. So, when they eventually catch on and start noticing the gulf between inflation and interest rates widening, as in my first graph, " so where does missing your target for five years running fit in? Why does anyone pretend that the Fed has credibility...OK a New Chair...Like Star Wars...A New Hope.. Is it enough?
Thank you for this clear analysis. Since for the past 4 decades, Fed-manipulated interest rates have been held to be "levers" in a policy machine, Warsh is apparently playing the role of Hippocrates, "first do no harm." Aside from some major shock, is there ever a reason for the Fed to move the policy rate? If one were a banker, jacking the IOER would be rewarding; would it encourage more reserves? Would that contract M1? M2?
That inflation was so quiet during the near decade of interest rates stuck at zero does sort of beg the question whether the Fed is doing any good by trying to move inflation around.
If deterrence only works when people believe the stick is real, then a “wisely cautious” hike is the one move that can’t actually deter anything — it just tells everyone the stick was never coming
Much of the news commentary about the mood of the public regarding Trump's economic performance seems to relate to consumers' memory of last year's prices on repeated purchases. This suggests that "inflation" may not actually be an economically objective phenomenon (although we have found many different ways "to measure" it). People seem to desire a Zero rate of change in annual average prices, particularly in food and fuel. Maybe the fear of "the zero boundary" is misplaced and the sacred 2% should be reconsidered. There is something unlovely about knowing the value of a Dollar will be halved in 36 years. Thank you for reminding us about 1947 prices.
Sure, fiscal reforms are important and can help stabilize inflation. However, the question is what monetary policy can achieve on its own. Accordingly, I assume that the IRFs in the top figure are plotted under the assumption that primary surpluses remain constant.
To prevent inflation from rising in the future following a monetary contraction, doesn't the model imply that the policy rate would need to decline much more rapidly than suggested by the figure and, in particular, fall below its initial level, converging back to the steady state in an oscillatory manner?
What would such a policy look like in practice? Would it imply that the Fed should begin lowering interest rates *before* it has achieved its inflation objective?
Great read, as always.
This is a simple policy, not an "optimal" policy. In this model like any other, give the Fed an objective and there is an "optimal" response to shocks. Whether the real Fed can implement that is a good question.
I have a question rather than a comment. If the central bank can do little about inflation, why does the central bank talking “tough” have any impact on inflation expectations? In your framework, if they talk tough and show a willingness to hike, wouldn’t that raise inflation expectations?
I’m not an economist but simplistically why doesn’t the Fed just suck money out of the economy via the banking system so prices can’t rise overall because there’s no money to pay higher prices? This would have a more general effect than interest rates, which can be very damaging for some individuals. Another mad idea is for the Fed to be able to add a Rainy Day Tax on income. The money is taken (which slows the economy) and is held in reserves to be released during the next slowdown as a stimulus?
Your second use is literally fiscal policy of the OG Keynesian sort.
Our Constitution does not allow the Fed to tax. Thank goodness.
A major mistake in statistical models is called the error of the omitted relevant variable. It’s nice to show graphs comparing one cause and an effect but its purpose should be limited to deciding whether to include that variable in a larger model. My analysis agrees with the result that higher real fed rates are associated with higher inflation two years hence but my 40 year model includes most of the usual suspects as variables too. And the results are good:R2 around .6 (0.7 if you omit the chaotic past 12 months). Besides the ornery behavior wrt Fed funds there are other surprises too. Hopefully I will have a paper by the fall.
Hey professor Cochrane,
Thanks for the article, it is a really interesting and undercovered point.
I wanted to make two questions/notes:
- One of the key assumptions is the long-run neutrality of money. I know that this is one of the most unquestionable ideas within classic economics. However, I wonder whether that is really applicable under interest rate targets. When the Central Bank alters or "manipulates" the interest rate, it distorts or alters real rates as well. Suddenly some investment projects could stop being profitable, or viceversa depending on the discount rate. In summary, the decision to change interest rates very likely has an effect on capital allocation decisions. It is also a fairly accepted fact that the stock of capital in an economy is a non-stationary series, it does have unit-roots, and it is not necessarily an ergodic process. In a less mathematical sense, it means that capital is long-lasting, and capital allocation decisions last very long (e.g. building a stock of housing, or building a chemical plant with a life of 50 years...). Those capital allocation decisions will therefore have long-run effects on the productive side of the economy, on the real side of things. Is really long-run neutrality true in this sense? I agree with the idea that units of measure do not change economic facts, but here it seems that the units of measure can distort or influence long-run decisions that do affect production in the long-run.
-My second point is a question about interest rate targets, monetarism and productivity improvements. From Warsh's reasoning, it can be understood that he believes that higher productivity lowers inflation (he is thinking in terms of M*V=P*Y). If Y grows faster so can M, however, is this equivalent to lower rates as Warsh implicitly assumes? Can we make money supply growth equivalent to lower interest rate?
Also, there are other economists and policy makers suggesting that in the medium-term, the policy rate should converge towards neutral, and if productivity is higher and demand for investment is higher so will be the neutral rate, therefore the Fed risks running loose policy if they cut. Which of these conflicting views do you think is right? Is it possible to conciliate these two totally opposed views?
Thank you. Always a pleasure to read your thoughts.
Christany
Really enjoyed this, Professor Cochrane — the "big stick" point is something I hadn't considered before. I'd been focused on whether Warsh sees today's inflation as more transitory than the consensus, not on whether the credibility behind his approach is actually backed by anything.
Apologies for the audacity, but I wrote a short piece on this from a different angle than yours and most of the analysis I've seen — https://www.linkedin.com/posts/artyom-shirinyan-1a6a62257_kevin-warsh-asked-the-market-to-speak-it-share-7490077572479868928-uMVO/?utm_source=share&utm_medium=member_android&rcm=ACoAAD9WnI4BzSof8W2IFMwU5IskF-FnLAv7KzQ .
Would genuinely value any feedback if you have a moment.