With so many important issues out there like who got canceled and what books are available at school libraries in San Antonio, throwing some monetary policy red meat to the base is at best a distraction.
Timely post! I've been looking for something like this.
Reading your description of the expectations theory, it sounds like it could be viewed as an extension of the hydraulic view rather than an alternative, i.e. the hydraulic view is right, but usually expectations move the market before the hydraulic mechanisms do. If the hydraulic view were entirely incorrect and the mechanisms had no effect, then there would be nothing for anyone to build expectations on, right?
If your title was an experiment in low expectation-setting, it worked because I didn’t think this was boring at all.
Like much else about our world, it’s both fascinating and terrifying to think the entire economy might be yet another figment of our imaginations; like government or the idea of God, it only works if we believe it works. On that note, I am happy that most people are in the hydraulic camp- it’s the only reason the expectation camp might actually be right.
Your opinion above of what Powell should be saying right now reminds me a lot of what I think you've been less explicitly suggesting the FDA/CDC should be saying about the pandemic. In other words...speak very frankly and clearly about what you're going to do and don't try to get clever about saying something that you hope will get the outcome you want without ticking anyone off or saying something controversial. I completely agree.
Greetings from Turkey, where President Erdogan is definitely making the central bank follow neo-Fischerian policies. Of course the low interest rates are making inflation worse rather than better, but they're making the currency collapse even faster than the domestic price level rises, so it makes for an amazingly cheap vacation.
Would just like to say that I was one of your readers who clicked on and actually read your monetary policy columns in Slate/Vox and that was a primary reason for me subscribing. And I'd much rather read about monetary/housing/transportation policy here than yet another column about Democrats' messaging problems (yes yes, I know it's necessary to read about that too... just that the balance seems to have tipped way too far in that direction).
Usually the posts on inflation here make sense, but this one is off the mark. The debate isn't about whether central banks *can* affect economic conditions by changing expectations. It's about whether central banks *should* try to implement monetary policy by changing expectations.
In basically any macroeconomic model, expectations about future interest rates affect current demand. If the central bank can make a credible announcement to cut rates in the future, then the economy will be stimulated in the present. When it comes time to actually cut rates, the central bank does so through "hydraulic" methods.
The disagreement among economists is whether the central bank's ability to control expectations is a scalpel or a machete. That is, if the central bank announces a 2% cut in the future, will people actually believe the cut will be exactly 2%?
The misconception about the expectations debate leads to some confusion throughout the rest of the piece.
1. Of course, everyone knew that long-term bond prices depend on expectations of tapering in 2013. The Fed was simply wrong about how much tapering the market expected, and they tapered more than markets thought, inadvertently causing the taper tantrum. This experience argues against implementation of monetary policy by expectations, not in favor of it.
2. The expectations-vs-hydraulics debate has nothing to do with whether central banks have to cause a recession in order to head off inflation. A well-calibrated interest rate hike in the present can suppress demand just the right amount today in the same way as a well-calibrated change in future expectations. The question is how easy it is to calibrate each intervention.
3. In the NZ example, the claim is that the central bank never actually needs to intervene as long as it announces an intervention ahead of time. Of course, if the central bank goes around announcing interventions it never implements, people will stop believing those announcements.
4. The Volcker disinflation again argues against expectations-driven monetary policy rather than in favor of it. During the 70s, the Fed repeatedly tried to assure markets that inflation was transitory and that it would hike rates when necessary. This didn't magically stop inflation because the Fed didn't actually succeed in changing expectations. Volcker had to rebuild credibility by actually raising rates.
nice article. and admirable humility. I told you immediately upon publishing your list of predictions around this time a year ago that your inflation priors were unlikely to be correct,
I think its admirable that you have changed your tune on this especially for the reason you listed: being a monetary policy dove has been one of your foundational views for over a decade now and has always served you well (I have generally been of the same persuasion)
I am not an expert, but I think you are conflating a few things here. One is how "traditional monetary policy" works (i.e. adjusting the federal funds rate) vs. how "quantitative easing" works.
My sense is the hydraulic mechanism of traditional monetary policy is pretty well understood. Higher federal funds rates increases borrowing costs to banks, so interest rates on other lending products go up, reducing demand for credit, reducing the money supply and slowing overall demand in the economy. [That the bond market (or stock market) doesn't move in lockstep with Fed actions doesn't really disprove this - capital markets "price in" all sorts of inputs including economic growth, inflation expectations, etc. so this is not a reliable indicator of anything here.]
QE is a much more novel tool, only having been invented in Japan 20 years ago and introduced in the US during the financial crisis, and is basically a way to get around the zero interest rate bound of traditional monetary policy. They hydraulics of this are much less well understood because there is so much less history here! And importantly unwinding QE has never been part of an inflation management strategy because the US, Japan and other markets where QE has been used have not seen robust inflation until now. I think experts disagree on 1) the extent to which QE has a big effect, 2) how that effect actually works, and 3) what would happen if it unwinds.
Beyond that distinction, I am also not clear how "coordinating expectations" can actually manage inflation. I agree keeping people confident that you will both manage inflation and support growth will keep economic activity high by providing the illusion (or reality) of stability, but that is a different problem than convincing people to slow down demand just a bit to ease off inflation without triggering a recession. Most economic actors are not making spending or borrowing decisions based on an expectation of 3% vs 2% GDP growth... they are looking at the cost of credit vs an opportunity in front of them. The hydraulic mechanism explains how to slow growth pretty straightforwardly, and I'm not sure I understand what the "expectations-led" mechanism would be.
One thing to note is that qualitative easing basically conjures money into existence; the fed buys stuff using money it creates. As a consequence, the amount of USD in existence (M0) has increased 83% since Feburary 2020 [1], which is kind of nuts! For months, people have been blaming various mechanisms for inflation (the price of rental cars if factored in! ports are slow! need more truck drivers!), but the amount of money in existence has nearly doubled in the last two years. It would be a major trick if there weren't significant inflation. (In fact, M0 given that M0 increased by ~40% per year, prices increasing by 5% per year seems pretty tame.) I don't see that much need in looking for hidden mechanisms for higher inflation.
I loved this post Matt! A very clear explanation of some tricky ground and I agreed with almost all of it.
However, one thing I found striking is that you seemed to be on the same page as Scott Sumner (whose book I'm currently reading) on almost everything until you got to fiscal policy. I was expecting you to take the Sumnerite position that monetary policy is so powerful that fiscal policy is rarely necessary or helpful. But instead you take the opposite view that the power of monetary policy makes it ok to overshoot on fiscal policy and then backtrack.
Is the issue here that you think expectations-based monetary policy is powerful most of the time but not in the most extreme circumstances? Or that it's powerful always in theory but political constraints occasionally prevent it from being powerful enough in practice? Or that you just aren't worried about debt levels so you see no real downside to throwing fiscal policy into the mix when a lot of stimulus is needed?
I took Macroeconomics in 1984 as an undergraduate econ major at Univ of Illinois from a prof named Case Sprenkle. Sprenkle showed up to lead the review for his final exam totally blotto and announced that "Macroeconomics is dead". I think what he meant was that all that M-1, M-2 and Laffer Curve (hydraulic) stuff wasn't predictive that the Fed can control the economy and inflation by levering interest rates and money supply.
Today it's clear, however, that levering expectations is the name of the game. That's where the modern MMT folks get it wrong. Sure, you can print as much money as you want, but if people believe that the government can print money to buy whatever it wants, why would they ever agree to pay taxes? Inflation is the economy killer and it's all about expectations.
I mean you could also argue that the continuation of Bernanke, Yellen and now Powell is that they preemptively stopped inflation, and now with Powell flipping during the pandemic we now have it.
Personally I think BOTH views are right, and the hydraulic view is what makes the perceptions view work.
I think of Fed policy as driving a car with an underpowered engine, brakes that sometimes don't engage right away, and a flakey speedometer.
To make things worse, you can't tell for sure if you're driving on a flat road, climbing a mountain, or descending a steep downhill. Also, you can't know what road grade you'll face around the next bend.
Now try to keep the car going 55 mph constantly and without error.
With so many important issues out there like who got canceled and what books are available at school libraries in San Antonio, throwing some monetary policy red meat to the base is at best a distraction.
The promise of “slow, boring” is why i signed up for this newsletter and I’m glad you’re keeping that promise.
Timely post! I've been looking for something like this.
Reading your description of the expectations theory, it sounds like it could be viewed as an extension of the hydraulic view rather than an alternative, i.e. the hydraulic view is right, but usually expectations move the market before the hydraulic mechanisms do. If the hydraulic view were entirely incorrect and the mechanisms had no effect, then there would be nothing for anyone to build expectations on, right?
If your title was an experiment in low expectation-setting, it worked because I didn’t think this was boring at all.
Like much else about our world, it’s both fascinating and terrifying to think the entire economy might be yet another figment of our imaginations; like government or the idea of God, it only works if we believe it works. On that note, I am happy that most people are in the hydraulic camp- it’s the only reason the expectation camp might actually be right.
Your opinion above of what Powell should be saying right now reminds me a lot of what I think you've been less explicitly suggesting the FDA/CDC should be saying about the pandemic. In other words...speak very frankly and clearly about what you're going to do and don't try to get clever about saying something that you hope will get the outcome you want without ticking anyone off or saying something controversial. I completely agree.
Greetings from Turkey, where President Erdogan is definitely making the central bank follow neo-Fischerian policies. Of course the low interest rates are making inflation worse rather than better, but they're making the currency collapse even faster than the domestic price level rises, so it makes for an amazingly cheap vacation.
Would just like to say that I was one of your readers who clicked on and actually read your monetary policy columns in Slate/Vox and that was a primary reason for me subscribing. And I'd much rather read about monetary/housing/transportation policy here than yet another column about Democrats' messaging problems (yes yes, I know it's necessary to read about that too... just that the balance seems to have tipped way too far in that direction).
Usually the posts on inflation here make sense, but this one is off the mark. The debate isn't about whether central banks *can* affect economic conditions by changing expectations. It's about whether central banks *should* try to implement monetary policy by changing expectations.
In basically any macroeconomic model, expectations about future interest rates affect current demand. If the central bank can make a credible announcement to cut rates in the future, then the economy will be stimulated in the present. When it comes time to actually cut rates, the central bank does so through "hydraulic" methods.
The disagreement among economists is whether the central bank's ability to control expectations is a scalpel or a machete. That is, if the central bank announces a 2% cut in the future, will people actually believe the cut will be exactly 2%?
The misconception about the expectations debate leads to some confusion throughout the rest of the piece.
1. Of course, everyone knew that long-term bond prices depend on expectations of tapering in 2013. The Fed was simply wrong about how much tapering the market expected, and they tapered more than markets thought, inadvertently causing the taper tantrum. This experience argues against implementation of monetary policy by expectations, not in favor of it.
2. The expectations-vs-hydraulics debate has nothing to do with whether central banks have to cause a recession in order to head off inflation. A well-calibrated interest rate hike in the present can suppress demand just the right amount today in the same way as a well-calibrated change in future expectations. The question is how easy it is to calibrate each intervention.
3. In the NZ example, the claim is that the central bank never actually needs to intervene as long as it announces an intervention ahead of time. Of course, if the central bank goes around announcing interventions it never implements, people will stop believing those announcements.
4. The Volcker disinflation again argues against expectations-driven monetary policy rather than in favor of it. During the 70s, the Fed repeatedly tried to assure markets that inflation was transitory and that it would hike rates when necessary. This didn't magically stop inflation because the Fed didn't actually succeed in changing expectations. Volcker had to rebuild credibility by actually raising rates.
nice article. and admirable humility. I told you immediately upon publishing your list of predictions around this time a year ago that your inflation priors were unlikely to be correct,
I think its admirable that you have changed your tune on this especially for the reason you listed: being a monetary policy dove has been one of your foundational views for over a decade now and has always served you well (I have generally been of the same persuasion)
keep up the good work
I am not an expert, but I think you are conflating a few things here. One is how "traditional monetary policy" works (i.e. adjusting the federal funds rate) vs. how "quantitative easing" works.
My sense is the hydraulic mechanism of traditional monetary policy is pretty well understood. Higher federal funds rates increases borrowing costs to banks, so interest rates on other lending products go up, reducing demand for credit, reducing the money supply and slowing overall demand in the economy. [That the bond market (or stock market) doesn't move in lockstep with Fed actions doesn't really disprove this - capital markets "price in" all sorts of inputs including economic growth, inflation expectations, etc. so this is not a reliable indicator of anything here.]
QE is a much more novel tool, only having been invented in Japan 20 years ago and introduced in the US during the financial crisis, and is basically a way to get around the zero interest rate bound of traditional monetary policy. They hydraulics of this are much less well understood because there is so much less history here! And importantly unwinding QE has never been part of an inflation management strategy because the US, Japan and other markets where QE has been used have not seen robust inflation until now. I think experts disagree on 1) the extent to which QE has a big effect, 2) how that effect actually works, and 3) what would happen if it unwinds.
Beyond that distinction, I am also not clear how "coordinating expectations" can actually manage inflation. I agree keeping people confident that you will both manage inflation and support growth will keep economic activity high by providing the illusion (or reality) of stability, but that is a different problem than convincing people to slow down demand just a bit to ease off inflation without triggering a recession. Most economic actors are not making spending or borrowing decisions based on an expectation of 3% vs 2% GDP growth... they are looking at the cost of credit vs an opportunity in front of them. The hydraulic mechanism explains how to slow growth pretty straightforwardly, and I'm not sure I understand what the "expectations-led" mechanism would be.
One thing to note is that qualitative easing basically conjures money into existence; the fed buys stuff using money it creates. As a consequence, the amount of USD in existence (M0) has increased 83% since Feburary 2020 [1], which is kind of nuts! For months, people have been blaming various mechanisms for inflation (the price of rental cars if factored in! ports are slow! need more truck drivers!), but the amount of money in existence has nearly doubled in the last two years. It would be a major trick if there weren't significant inflation. (In fact, M0 given that M0 increased by ~40% per year, prices increasing by 5% per year seems pretty tame.) I don't see that much need in looking for hidden mechanisms for higher inflation.
[1] https://fred.stlouisfed.org/series/BOGMBASE
I loved this post Matt! A very clear explanation of some tricky ground and I agreed with almost all of it.
However, one thing I found striking is that you seemed to be on the same page as Scott Sumner (whose book I'm currently reading) on almost everything until you got to fiscal policy. I was expecting you to take the Sumnerite position that monetary policy is so powerful that fiscal policy is rarely necessary or helpful. But instead you take the opposite view that the power of monetary policy makes it ok to overshoot on fiscal policy and then backtrack.
Is the issue here that you think expectations-based monetary policy is powerful most of the time but not in the most extreme circumstances? Or that it's powerful always in theory but political constraints occasionally prevent it from being powerful enough in practice? Or that you just aren't worried about debt levels so you see no real downside to throwing fiscal policy into the mix when a lot of stimulus is needed?
I took Macroeconomics in 1984 as an undergraduate econ major at Univ of Illinois from a prof named Case Sprenkle. Sprenkle showed up to lead the review for his final exam totally blotto and announced that "Macroeconomics is dead". I think what he meant was that all that M-1, M-2 and Laffer Curve (hydraulic) stuff wasn't predictive that the Fed can control the economy and inflation by levering interest rates and money supply.
Today it's clear, however, that levering expectations is the name of the game. That's where the modern MMT folks get it wrong. Sure, you can print as much money as you want, but if people believe that the government can print money to buy whatever it wants, why would they ever agree to pay taxes? Inflation is the economy killer and it's all about expectations.
I mean you could also argue that the continuation of Bernanke, Yellen and now Powell is that they preemptively stopped inflation, and now with Powell flipping during the pandemic we now have it.
Personally I think BOTH views are right, and the hydraulic view is what makes the perceptions view work.
Terrible news I thought CPI numbers were going to be 6.9% Y/Y.
I think of Fed policy as driving a car with an underpowered engine, brakes that sometimes don't engage right away, and a flakey speedometer.
To make things worse, you can't tell for sure if you're driving on a flat road, climbing a mountain, or descending a steep downhill. Also, you can't know what road grade you'll face around the next bend.
Now try to keep the car going 55 mph constantly and without error.