The proposal, starting with comments from Senator Ted Cruz, that the Congress force the Fed to stop paying interest on reserves is having more legs than I thought it would.
I agree that stopping IOR payments would be inflationary. As for explaining why, I think it tracks back to explaining why monetarism (as a predictive economic theory) broke down in the first place.
Monetarist theory relies on a friction between cash and Treasury bonds. Cash pays no interest but is perfectly illiquid. Conversely, treasury bonds pay interest but are somewhat illiquid. The tradeoff between liquidity and interest results in a well-defined money demand function. This (alongside long-run monetary neutrality) is sufficient to yield the usual monetarist predictions.
This friction has broken down over the decades due to (1) increased Treasury market liquidity, and (2) interest paying cash. The latter includes interest on reserves, but also (e.g.) earlier regulatory reforms (I believe Reg Q?) allowing banks to pay market interest rates on deposits.
Absent the friction between cash and bonds, the notion of "money demand" breaks down. The Fed might buy long-term Treasuries in exchange for reserves, and this will increase a money supply measure like M2, but "M2 money demand" (now ill-defined) will move to offset it.
Because of this, the usual monetarist mechanism (i.e., given a change in the nominal money supply, inflation equates real money supplied with real money demanded) is no longer operable.
Removing IOR makes cash and Treasuries less substitutable. The (lower than typical) levels of Treasury market liquidity since the pandemic also do not help. It's as though your truck is stuck in a muddy ditch, your foot is clamped down on the accelerator, and suddenly there's traction.
At least, this is how I've boiled it down to people in politics. I hope this is not too incorrect!
You can have a well-defined money demand function if you broaden your definition of money and weight assets by their liquidity. Divisia M4 includes short term treasuries, and has a money demand function that allows us to make sense of many of the episodes monetarism "fails" to explain.
Would you agree (as Berkshire holdings seem to) or disagree that any treasuries out >1 yr are quite risky? Including TIPS, as there is a risk of CPI definition erosion, or just reneging?
The Vissing-Jorgensen Jackson Hole paper suggests the IOR rate should be lower than SOFR by the convenience yield on Treasuries (70bps?), if I'm understanding correctly. This would certainly result in lower reserves, but they'd still but high by pre-GFC standards I imagine.
Senator Cruz is attempting to find a source of revenue for the government so that the Republican tax cuts can meet a necessary first order condition imposed by law. Because the FRB has historically transferred its annual operating surplus to the Treasury, it appears that the FRB is one such source of revenue. It is attractive because, in the senator's view, that stream of revenue is available without raising taxes or cutting expenditures. Otherwise, why do it?
Commercial banks hold reserves in two forms: first, as vault reserves in the form of U.S. currency; second, as deposits with the regional Federal Reserve Banks in the form of U.S. currency. Hypothetically, commercial banks could hold reserves in the form of Treasury bills but for fact that the supply of Treasury bills is constrained and the cost to the banks in terms of managing such assets for the purpose of settling inter-bank credits and debits on a daily basis at values that fluctuating minute by minute would overwhelm the banks' systems, whereas U.S.currency has a fixed an invariant nominal value that poses no such problems for settlement.
The next question to be addressed by the senator is the challenge that Federal Reserve Bank of New York's trading desk is faced with minute by minute during the trading day, i.e. 24/7, namely the maintenance of the effective Fed Funds rate at the FOMC's target Federal Funds Rate. This task is made easier by paying interest on deposits held at the Federal Reserve regional banks, sometimes referred to as "reserves". Absent payment of interest on reserves, the trading desk at the FRBNY finds that it cannot support the FOMC's target FFR, according to the FRBNY. Monetary policy then becomes significantly more difficult to transmit to the financial markets via the interest rate channel. If monetary policy cannot be effectively transmitted through the interest rate, then direct manipulation of the national money supply may have to be resorted to.
These appear to be the arguments set forth in response to the senator's proposed legislative initiative. Whether those arguments will cut ice with the Senate remains to be seen.
As to the question of whether or not the FOMC will be forced to taper its holdings of Treasury securities, mortgage collateralized debt securities, and various other private issue debentures and notes if interest is not paid on reserves whether by FOMC policy or by virtue of federal legislation one has to look at demand for interest-paying deposits at the regional Federal Reserve Banks. This question hinges on the convenience yield obtained by holding bank reserves in the form of deposits held at the regional Federal Reserve Banks, esp. the FRBNY. This being testable, an opinion is of limited value to the discussion.
In essence the senator's testing of his colleagues' appetite for innovation has sparked a fundamental challenge to the transmission of monetary policy via the interest rate channel.
I don't know if doing away with interest on reserves is inflationary but it probably represents opportunity cost to banks. They complained for years that they received no interest on reserves and finally got their way. I think now it is another revenue stream that they count on and may encourage more risk. I would like to know what the connection is between IOR and risk. Thanks.
It is in places somewhat technical in nature, but, given that the monetary policy is highly technical in view of the means required to convert policy to actual behavior of the players in this field, that is to be expected. Nevertheless, Peter Ireland does a yeoman's job of explaining the subject aptedly described by the title of the paper to a largely non-technical audience.
In Ireland's telling, and one should note the publication date of the second calendar quarter of 2019, it is the payment of interest on excess reserves that poses a risk to the financial system. Ireland puts forward M. Goodfriend's proposals for changing policy to minimize those risks.
As to whether or not eliminating interest on reserves would have an inflationary effect, one would have to examine the behavior of the commercial banks' propensity to lend in alternatively (a) an environment in which the central bank offers IOR, and (b) an environment in which the central bank does not offer IOR. Ireland maintains that when the central bank commenced paying IOER (interest on excess reserves), the demand curve for reserves shifted to the right (i.e., demand increased), and, simultaneously, the supply curve for reserves also shifted to the right (i.e., supply increased). Now, consider the case where Congress amends the Federal Reserve Act to remove the clauses that permit the Federal Reserve Bank System to pay IOR. Logically, we would expect that on the central bank stopping IOR (IOER), the demand curve for reserves will shift to the left (demand decreases), and the supply curve will also shift to the left (supply decreases).
Now, if a commercial bank cannot obtain IOR, or IOER, its idle balances (invested in FRB reserves) will seek a new 'home', if one can be found, to make up for the loss of interest earned on its reserve balances held at the FRB's regional banks. Vault reserves don't supply that service, but increasing lending possibly might. If credit availability increases as a result of the loss of IOR (IOER), then it is likely that aggregate demand will increase by some fraction, less than one but greater than zero, when aggregate supply is restrained by the lag time required to resupply inventories. A temporary rise in the rate of inflation (increase in the general price level, P(t)) might result. This appears to be consistent with central bank theory that states that counter-cyclical policy during credit expansions is to have the central bank attract deposits from commercial banks in order to reduce the availability of credit facilities to firms and households.
Whether this addresses your query or not, I encourage you to peruse Peter Ireland's article cited above, and the many references he tallies at the end of the paper. Also, don't take my word for gospel--because it isn't. Ask others, and consult the libraries for more information.
Peter is a great economist, but I disagree here. The heart of the monetarist position is that the composition, not the quantity, of government debt matters. Exchanging $1 trillion of one month treasury debt for $1 trillion of reserves (overnight government debt) that pay almost exactly the same interest, is inflationary. That QE had no effect on inflation, but issuing more debt did seems to prove the contrary.
I'm not sure that the QE link to changes in the general price level actually disproves the monetarist theory, given the economic conditions that led to QE in the first place. I'll leave it to Peter Ireland to argue the merits of his point of view.
The FRBNY trading desk has been reducing its asset balances at an average rate of $1.375B per day since 9/07/2022. Regression curve: A(t) = A(0) - m.t, where A(0) = 4.864 trillions, m = -1.3752 billions/day t=0 at 9/07/2022. t = 0, 1, 2, ... . [Standard error of estimate of m = 22.16 millions (t-stat: 62.05). Goodness of fit 0.964. F-stat: 3849.7 Deg. of freedom 144. Observations: 146 weekly observations. Series: D2WATAL. ] This is a linear trend line. Curvature is effectively zero. Graphical presentation of the time series D2WATAL alongside PCE inflation, PCEPI y/y %change, and Fed Funds Effective Rate, FEDFUNDS, is found at https://fred.stlouisfed.org/graph/?g=1JKw6 Only the data from 9/7/2022 to 6/10/2025 was used for the regression analysis, but the ramp up and the ramp down of D2WATAL relative to PCEPI y/y %change is of interest.
The effect of QE1, QE2, and QE3 on D2WATAL can be seen clearly. The y/y %change in PCEPI can be seen in the sharp upswing of that measure in H2 of 2009 corresponding to initiation of QE1 which mirrors (except in scale) the sharp rise in y/y %changed in PCEPI in 2020-21 even down to the lag time between the injection of liquidity and the outbreak of inflation (from the disinflationary trend during the economic crises). The subsequent episodes of QE2 and QE3 correspond to the effort by FOMC to avoid disflationary tendencies. That QE2 and QE3 had only marginal and not consistent effects on the inflation rate can likely be attributed to modest objectives set for QE2 and QE3, as noted by several economists active in this field. In contrast the fiscal 'helicopter dump' in 2020 and 2021 was more direct in expanding the money supply via base money increases and the inflation rate was thereby more pronounced. That the FTPL conditions for avoiding increased inflationary expectations were absent in 2020 and 2021 (i.e., credible commitment by the government--Congress--to run future primary surpluses to soak up the excess indebtedness incurred by the shorterm fiscal stimulus) supports the cautions around using the fiscal channel to counter an economic downturn. The concommitant monetary stimulus by the FOMC to contain a budding financial crisis in early 2020 can be viewed as a contributing factor to the sharp upswing in PCEPI during that period.
Fiscal stimulus without commitment in conjunction with an expansionary monetary policy by the central bank (when central bank standard operating procedure prior to this event would entail tightening monetary policy to counteract fiscal stimulus) makes isolation of the two channels of monetary policy challenging.
I am not at all certain that QE1, QE2, and QE3 had no effect on the rate of inflation. Those quantitative easing tranches were intended to counteract a disinflationary inclination in the economy and to improve the availability of affordable credit to a broad spectrum of creditworthy firms and institutions subsequent to the failure of Lehman Bros. (i.e., a precautionary monetary policy). The contemporary active discussion around the zero lower bound (ZLB) and negative interest rates indicates that boosting the then-current rate of inflation was an objective of monetary policy during the period that the FOMC was effecting quantitative easing.
But, as I indicated earlier, I leave it to Peter Ireland to argue his point of view on the connection between IOR/IOER and the general price inflation PCEPI y/y %change.
How does the Fed make IOR payments (which are lower than the Treasury coupons holdings)? Printing money?
In Argentina the Central Bank used to have LEBACs (CB Bills) with very high interest rates. In 2023 (before new administration), the Central Bank issued more money to pay LEBAC interest payments than to finance the Treasury. Financial dominance > Fiscal dominance.
I wholeheartedly agree that the evidence for lower interest rates causing inflation is murky. Conversely, evidence that raising rates cures inflation is murky as well. In fact, there is zero evidence in the data for either. The only clear evidence from the 1970s is that raising rates too high above the inflation rate for no reason can be a precursor to an inflation spike.
"So eliminating interest on reserves is equivalent to commanding “set the interest rate target to zero.” Given President Trump’s desire to see lower interest rates, maybe that’s what’s at work." Exactly my first thought. It'll force the FED's hand to set interest rates close to zero. My second thought: Maybe the FED could keep interest rates high by accepting reverse Repos as reserves? So the whole policy seems futile (except it may give the federal government the fiscal headroom it is currently seeking.) On the inflation front: If the government uses the additional fiscal headroom afforded by such measure to spend more, that additional spending will generate inflation, not the low interest rate.
I wonder what difference it would make if the first thing we do in discussions about the Fed is acknowledge that the commercial banking industry has been nationalized.
Personally, I think that would be clarifying. Maybe it wouldn't make a difference. I don't know. But it would be clarifying.
it is true though, that some of the reward is private, but the government decides to whom those rewards flow
try acquiring 9+% of a bank or holding company
but you're right about one thing: if I want to acquire a small amount of bank or holding company stock, the government will allow me to do so unmolested, and I will potentially make a gain that is slightly better than lousy
I didn’t suggest there wasn’t crony capitalism involved.
My point is in key ways it is worse than you suggest with your “nationalized” claim, since taxpayers absorb most of the losses but private entities get most of the profits.
Why political leaders seem to prefer the magic wand approach to solving issues is a mystery to me. Maybe they’re just lazy. Focusing on one specific thread in a multi-threaded discipline such as economics, ignores the bigger picture, and can cause serious imbalances if implemented. Seeing the Big Picture, though rare among policymakers, is a desirable goal. A picture is worth a thousand words. But try sticking to a few hundred if you please….
I will push back, if I may, against your criticism of the "old way" of managing bank reserves, when there was no interest paid. (I am not 100% sure about all of this. I'd be interested in your thoughts.) Today, the Fed Funds rate is an entirely administered rate arbitrarily determined by the Fed. It bears zero relationship to the actual demand for and supply of credit in our economy.
Prior to 2008, and especially since 2020, Fed Funds was at least a quasi-market interest rate, driven by the economy’s underlying demand for and supply of credit. The policy mechanism was not the setting of interest rates per se, but the setting the quantity of bank reserves. Banks with insufficient reserves at the end of the reporting period could borrow reserves from those with excess reserves. If banks needed to bid above the target Fed Funds rate for reserves, the Fed could buy securities, thereby introducing more reserves into the system and tamping down the Fed Funds rate. In contrast, today’s interest rate targets are purely arbitrary shots in the dark.
Also, I should point out that banks face a spiderweb of capital and liquidity regulations that will severely complicate any end to interest on reserves. As things stand now, I think banks would be more inclined to shrink balance sheets than buy longer-dated Treasuries.
Also, remember that while an individual bank can reduce reserves, the "system as a whole" cannot reduce reserves unless the Fed sells securities which I am not sure "the system" would be equipped to absorb. (I think this may be what you were driving at, but I wasn't sure.)
Lastly, I am astounded that our populist legislators are only now beginning to appreciate this issue.
Good points, often stated. But the NY fed adjusted reserves every day so that the next day it would hit the interest rate target. Market determined prices facing a fixed supply lasted one day. Did that really matter?
Also, I want to push back against economists who disparage monetarism and Friedman. The rejection of monetarism explains why every orthodox economist was dead wrong about the inflation outlook in 2021. We simply printed too much money. End of story.
The US also printed too much money in QE, leading to no inflation. In 2021-2022 it printed that money to finance deficits, and got a lot of inflation. It's not just the money. Using money to buy bonds seems to have no effect per se if it's not financing deficits. And most of Friedman's episodes involved printing money to finance deficits.
Why am I more and more feeling like I did as a kid when visiting my first carnival, and watching the other people play the shill games and expecting to win?
What's wrong with the Fed neither requiring nor paying interest on reserves. If prudential regulators insist on "reserves" in addition to capital bank can just buy Tbills.
“The good old days were awful. Ample reserves that pay market interest are a great innovation. They cost nothing. In fact they save the government money, as reserves typically pay less interest than longer maturity debt. The banks would otherwise hold directly the same treasury debt that the Fed holds, and there would be no rebate.”
The quoted statement above is fantastic.
It is, I think the key point.
I’m quite certain that I know WAY more economics than the average person or politician (undergrad econ major, plus have followed several econ podcasts for many years and now a few Substacks with great interest). I know something about trade flows, enough to know that both sides in the kerfuffle about what Trump may or may not do with tariffs are either being dishonest or clueless much of the time.
I say all this because despite my relatively high econ knowledge, I was utterly unaware of this point about the innovation that ample reserves that pay market rate interest are.
I suspect most people are equally unaware.
Hopefully JHC will write another, shorter post where this lede is not buried. 😏
A small step could to pay the sofr rate (which is lower than ior much of the time). Alternative policy instruments like unwinding qe could be used to tighten monetary policy instead of using ior for this purpose.
As regards fed losses, they could reduce this a bit by buying more or just tips bonds, which would be less subject to losses.
Excellent post
I agree that stopping IOR payments would be inflationary. As for explaining why, I think it tracks back to explaining why monetarism (as a predictive economic theory) broke down in the first place.
Monetarist theory relies on a friction between cash and Treasury bonds. Cash pays no interest but is perfectly illiquid. Conversely, treasury bonds pay interest but are somewhat illiquid. The tradeoff between liquidity and interest results in a well-defined money demand function. This (alongside long-run monetary neutrality) is sufficient to yield the usual monetarist predictions.
This friction has broken down over the decades due to (1) increased Treasury market liquidity, and (2) interest paying cash. The latter includes interest on reserves, but also (e.g.) earlier regulatory reforms (I believe Reg Q?) allowing banks to pay market interest rates on deposits.
Absent the friction between cash and bonds, the notion of "money demand" breaks down. The Fed might buy long-term Treasuries in exchange for reserves, and this will increase a money supply measure like M2, but "M2 money demand" (now ill-defined) will move to offset it.
Because of this, the usual monetarist mechanism (i.e., given a change in the nominal money supply, inflation equates real money supplied with real money demanded) is no longer operable.
Removing IOR makes cash and Treasuries less substitutable. The (lower than typical) levels of Treasury market liquidity since the pandemic also do not help. It's as though your truck is stuck in a muddy ditch, your foot is clamped down on the accelerator, and suddenly there's traction.
At least, this is how I've boiled it down to people in politics. I hope this is not too incorrect!
You can have a well-defined money demand function if you broaden your definition of money and weight assets by their liquidity. Divisia M4 includes short term treasuries, and has a money demand function that allows us to make sense of many of the episodes monetarism "fails" to explain.
Would you agree (as Berkshire holdings seem to) or disagree that any treasuries out >1 yr are quite risky? Including TIPS, as there is a risk of CPI definition erosion, or just reneging?
I don’t claim to know the answers to your questions, but TIPS are a VERY different thing than all other treasuries, and any risks are quite different.
The Vissing-Jorgensen Jackson Hole paper suggests the IOR rate should be lower than SOFR by the convenience yield on Treasuries (70bps?), if I'm understanding correctly. This would certainly result in lower reserves, but they'd still but high by pre-GFC standards I imagine.
Senator Cruz is attempting to find a source of revenue for the government so that the Republican tax cuts can meet a necessary first order condition imposed by law. Because the FRB has historically transferred its annual operating surplus to the Treasury, it appears that the FRB is one such source of revenue. It is attractive because, in the senator's view, that stream of revenue is available without raising taxes or cutting expenditures. Otherwise, why do it?
Commercial banks hold reserves in two forms: first, as vault reserves in the form of U.S. currency; second, as deposits with the regional Federal Reserve Banks in the form of U.S. currency. Hypothetically, commercial banks could hold reserves in the form of Treasury bills but for fact that the supply of Treasury bills is constrained and the cost to the banks in terms of managing such assets for the purpose of settling inter-bank credits and debits on a daily basis at values that fluctuating minute by minute would overwhelm the banks' systems, whereas U.S.currency has a fixed an invariant nominal value that poses no such problems for settlement.
The next question to be addressed by the senator is the challenge that Federal Reserve Bank of New York's trading desk is faced with minute by minute during the trading day, i.e. 24/7, namely the maintenance of the effective Fed Funds rate at the FOMC's target Federal Funds Rate. This task is made easier by paying interest on deposits held at the Federal Reserve regional banks, sometimes referred to as "reserves". Absent payment of interest on reserves, the trading desk at the FRBNY finds that it cannot support the FOMC's target FFR, according to the FRBNY. Monetary policy then becomes significantly more difficult to transmit to the financial markets via the interest rate channel. If monetary policy cannot be effectively transmitted through the interest rate, then direct manipulation of the national money supply may have to be resorted to.
These appear to be the arguments set forth in response to the senator's proposed legislative initiative. Whether those arguments will cut ice with the Senate remains to be seen.
As to the question of whether or not the FOMC will be forced to taper its holdings of Treasury securities, mortgage collateralized debt securities, and various other private issue debentures and notes if interest is not paid on reserves whether by FOMC policy or by virtue of federal legislation one has to look at demand for interest-paying deposits at the regional Federal Reserve Banks. This question hinges on the convenience yield obtained by holding bank reserves in the form of deposits held at the regional Federal Reserve Banks, esp. the FRBNY. This being testable, an opinion is of limited value to the discussion.
In essence the senator's testing of his colleagues' appetite for innovation has sparked a fundamental challenge to the transmission of monetary policy via the interest rate channel.
I don't know if doing away with interest on reserves is inflationary but it probably represents opportunity cost to banks. They complained for years that they received no interest on reserves and finally got their way. I think now it is another revenue stream that they count on and may encourage more risk. I would like to know what the connection is between IOR and risk. Thanks.
Answer from an economist's point of view (being that of one who avers to the 'monetarist' point of view) can be had from Peter Ireland, Ph.D. (econ.).
Ireland, Peter, "Interest on Reserves: History and Rationale, Complications and Risks", CATO Journal Spring/Summer 2019. Free access online.
URL: https://www.cato.org/cato-journal/spring/summer-2019/interest-reserves-history-rationale-complications-risks . Twenty-nine references. PDF version available for download from the CATO website (see URL).
It is in places somewhat technical in nature, but, given that the monetary policy is highly technical in view of the means required to convert policy to actual behavior of the players in this field, that is to be expected. Nevertheless, Peter Ireland does a yeoman's job of explaining the subject aptedly described by the title of the paper to a largely non-technical audience.
In Ireland's telling, and one should note the publication date of the second calendar quarter of 2019, it is the payment of interest on excess reserves that poses a risk to the financial system. Ireland puts forward M. Goodfriend's proposals for changing policy to minimize those risks.
As to whether or not eliminating interest on reserves would have an inflationary effect, one would have to examine the behavior of the commercial banks' propensity to lend in alternatively (a) an environment in which the central bank offers IOR, and (b) an environment in which the central bank does not offer IOR. Ireland maintains that when the central bank commenced paying IOER (interest on excess reserves), the demand curve for reserves shifted to the right (i.e., demand increased), and, simultaneously, the supply curve for reserves also shifted to the right (i.e., supply increased). Now, consider the case where Congress amends the Federal Reserve Act to remove the clauses that permit the Federal Reserve Bank System to pay IOR. Logically, we would expect that on the central bank stopping IOR (IOER), the demand curve for reserves will shift to the left (demand decreases), and the supply curve will also shift to the left (supply decreases).
Now, if a commercial bank cannot obtain IOR, or IOER, its idle balances (invested in FRB reserves) will seek a new 'home', if one can be found, to make up for the loss of interest earned on its reserve balances held at the FRB's regional banks. Vault reserves don't supply that service, but increasing lending possibly might. If credit availability increases as a result of the loss of IOR (IOER), then it is likely that aggregate demand will increase by some fraction, less than one but greater than zero, when aggregate supply is restrained by the lag time required to resupply inventories. A temporary rise in the rate of inflation (increase in the general price level, P(t)) might result. This appears to be consistent with central bank theory that states that counter-cyclical policy during credit expansions is to have the central bank attract deposits from commercial banks in order to reduce the availability of credit facilities to firms and households.
Whether this addresses your query or not, I encourage you to peruse Peter Ireland's article cited above, and the many references he tallies at the end of the paper. Also, don't take my word for gospel--because it isn't. Ask others, and consult the libraries for more information.
Peter is a great economist, but I disagree here. The heart of the monetarist position is that the composition, not the quantity, of government debt matters. Exchanging $1 trillion of one month treasury debt for $1 trillion of reserves (overnight government debt) that pay almost exactly the same interest, is inflationary. That QE had no effect on inflation, but issuing more debt did seems to prove the contrary.
I'm not sure that the QE link to changes in the general price level actually disproves the monetarist theory, given the economic conditions that led to QE in the first place. I'll leave it to Peter Ireland to argue the merits of his point of view.
The FRBNY trading desk has been reducing its asset balances at an average rate of $1.375B per day since 9/07/2022. Regression curve: A(t) = A(0) - m.t, where A(0) = 4.864 trillions, m = -1.3752 billions/day t=0 at 9/07/2022. t = 0, 1, 2, ... . [Standard error of estimate of m = 22.16 millions (t-stat: 62.05). Goodness of fit 0.964. F-stat: 3849.7 Deg. of freedom 144. Observations: 146 weekly observations. Series: D2WATAL. ] This is a linear trend line. Curvature is effectively zero. Graphical presentation of the time series D2WATAL alongside PCE inflation, PCEPI y/y %change, and Fed Funds Effective Rate, FEDFUNDS, is found at https://fred.stlouisfed.org/graph/?g=1JKw6 Only the data from 9/7/2022 to 6/10/2025 was used for the regression analysis, but the ramp up and the ramp down of D2WATAL relative to PCEPI y/y %change is of interest.
The effect of QE1, QE2, and QE3 on D2WATAL can be seen clearly. The y/y %change in PCEPI can be seen in the sharp upswing of that measure in H2 of 2009 corresponding to initiation of QE1 which mirrors (except in scale) the sharp rise in y/y %changed in PCEPI in 2020-21 even down to the lag time between the injection of liquidity and the outbreak of inflation (from the disinflationary trend during the economic crises). The subsequent episodes of QE2 and QE3 correspond to the effort by FOMC to avoid disflationary tendencies. That QE2 and QE3 had only marginal and not consistent effects on the inflation rate can likely be attributed to modest objectives set for QE2 and QE3, as noted by several economists active in this field. In contrast the fiscal 'helicopter dump' in 2020 and 2021 was more direct in expanding the money supply via base money increases and the inflation rate was thereby more pronounced. That the FTPL conditions for avoiding increased inflationary expectations were absent in 2020 and 2021 (i.e., credible commitment by the government--Congress--to run future primary surpluses to soak up the excess indebtedness incurred by the shorterm fiscal stimulus) supports the cautions around using the fiscal channel to counter an economic downturn. The concommitant monetary stimulus by the FOMC to contain a budding financial crisis in early 2020 can be viewed as a contributing factor to the sharp upswing in PCEPI during that period.
Fiscal stimulus without commitment in conjunction with an expansionary monetary policy by the central bank (when central bank standard operating procedure prior to this event would entail tightening monetary policy to counteract fiscal stimulus) makes isolation of the two channels of monetary policy challenging.
I am not at all certain that QE1, QE2, and QE3 had no effect on the rate of inflation. Those quantitative easing tranches were intended to counteract a disinflationary inclination in the economy and to improve the availability of affordable credit to a broad spectrum of creditworthy firms and institutions subsequent to the failure of Lehman Bros. (i.e., a precautionary monetary policy). The contemporary active discussion around the zero lower bound (ZLB) and negative interest rates indicates that boosting the then-current rate of inflation was an objective of monetary policy during the period that the FOMC was effecting quantitative easing.
But, as I indicated earlier, I leave it to Peter Ireland to argue his point of view on the connection between IOR/IOER and the general price inflation PCEPI y/y %change.
Thank you for this post!
How does the Fed make IOR payments (which are lower than the Treasury coupons holdings)? Printing money?
In Argentina the Central Bank used to have LEBACs (CB Bills) with very high interest rates. In 2023 (before new administration), the Central Bank issued more money to pay LEBAC interest payments than to finance the Treasury. Financial dominance > Fiscal dominance.
I wholeheartedly agree that the evidence for lower interest rates causing inflation is murky. Conversely, evidence that raising rates cures inflation is murky as well. In fact, there is zero evidence in the data for either. The only clear evidence from the 1970s is that raising rates too high above the inflation rate for no reason can be a precursor to an inflation spike.
"So eliminating interest on reserves is equivalent to commanding “set the interest rate target to zero.” Given President Trump’s desire to see lower interest rates, maybe that’s what’s at work." Exactly my first thought. It'll force the FED's hand to set interest rates close to zero. My second thought: Maybe the FED could keep interest rates high by accepting reverse Repos as reserves? So the whole policy seems futile (except it may give the federal government the fiscal headroom it is currently seeking.) On the inflation front: If the government uses the additional fiscal headroom afforded by such measure to spend more, that additional spending will generate inflation, not the low interest rate.
I wonder what difference it would make if the first thing we do in discussions about the Fed is acknowledge that the commercial banking industry has been nationalized.
Personally, I think that would be clarifying. Maybe it wouldn't make a difference. I don't know. But it would be clarifying.
But it has not.
Risks have been nationalized.
But rewards are still mostly private.
In that sense it is much worse than your claim.
oh yes it has
it is true though, that some of the reward is private, but the government decides to whom those rewards flow
try acquiring 9+% of a bank or holding company
but you're right about one thing: if I want to acquire a small amount of bank or holding company stock, the government will allow me to do so unmolested, and I will potentially make a gain that is slightly better than lousy
I didn’t suggest there wasn’t crony capitalism involved.
My point is in key ways it is worse than you suggest with your “nationalized” claim, since taxpayers absorb most of the losses but private entities get most of the profits.
Why political leaders seem to prefer the magic wand approach to solving issues is a mystery to me. Maybe they’re just lazy. Focusing on one specific thread in a multi-threaded discipline such as economics, ignores the bigger picture, and can cause serious imbalances if implemented. Seeing the Big Picture, though rare among policymakers, is a desirable goal. A picture is worth a thousand words. But try sticking to a few hundred if you please….
I will push back, if I may, against your criticism of the "old way" of managing bank reserves, when there was no interest paid. (I am not 100% sure about all of this. I'd be interested in your thoughts.) Today, the Fed Funds rate is an entirely administered rate arbitrarily determined by the Fed. It bears zero relationship to the actual demand for and supply of credit in our economy.
Prior to 2008, and especially since 2020, Fed Funds was at least a quasi-market interest rate, driven by the economy’s underlying demand for and supply of credit. The policy mechanism was not the setting of interest rates per se, but the setting the quantity of bank reserves. Banks with insufficient reserves at the end of the reporting period could borrow reserves from those with excess reserves. If banks needed to bid above the target Fed Funds rate for reserves, the Fed could buy securities, thereby introducing more reserves into the system and tamping down the Fed Funds rate. In contrast, today’s interest rate targets are purely arbitrary shots in the dark.
Also, I should point out that banks face a spiderweb of capital and liquidity regulations that will severely complicate any end to interest on reserves. As things stand now, I think banks would be more inclined to shrink balance sheets than buy longer-dated Treasuries.
Also, remember that while an individual bank can reduce reserves, the "system as a whole" cannot reduce reserves unless the Fed sells securities which I am not sure "the system" would be equipped to absorb. (I think this may be what you were driving at, but I wasn't sure.)
Lastly, I am astounded that our populist legislators are only now beginning to appreciate this issue.
See my substack posts https://charles72f.substack.com/p/auctions-and-tapers
https://charles72f.substack.com/p/the-looming-liquidity-crisis-and
https://charles72f.substack.com/p/law-and-order-fdic
Good points, often stated. But the NY fed adjusted reserves every day so that the next day it would hit the interest rate target. Market determined prices facing a fixed supply lasted one day. Did that really matter?
Also, I want to push back against economists who disparage monetarism and Friedman. The rejection of monetarism explains why every orthodox economist was dead wrong about the inflation outlook in 2021. We simply printed too much money. End of story.
https://charles72f.substack.com/p/aint-nothin-but-a-party
The US also printed too much money in QE, leading to no inflation. In 2021-2022 it printed that money to finance deficits, and got a lot of inflation. It's not just the money. Using money to buy bonds seems to have no effect per se if it's not financing deficits. And most of Friedman's episodes involved printing money to finance deficits.
Surely it depends where the money goes? Didn’t we have massive asset price inflation during QE?
Why am I more and more feeling like I did as a kid when visiting my first carnival, and watching the other people play the shill games and expecting to win?
What's wrong with the Fed neither requiring nor paying interest on reserves. If prudential regulators insist on "reserves" in addition to capital bank can just buy Tbills.
“The good old days were awful. Ample reserves that pay market interest are a great innovation. They cost nothing. In fact they save the government money, as reserves typically pay less interest than longer maturity debt. The banks would otherwise hold directly the same treasury debt that the Fed holds, and there would be no rebate.”
The quoted statement above is fantastic.
It is, I think the key point.
I’m quite certain that I know WAY more economics than the average person or politician (undergrad econ major, plus have followed several econ podcasts for many years and now a few Substacks with great interest). I know something about trade flows, enough to know that both sides in the kerfuffle about what Trump may or may not do with tariffs are either being dishonest or clueless much of the time.
I say all this because despite my relatively high econ knowledge, I was utterly unaware of this point about the innovation that ample reserves that pay market rate interest are.
I suspect most people are equally unaware.
Hopefully JHC will write another, shorter post where this lede is not buried. 😏
A small step could to pay the sofr rate (which is lower than ior much of the time). Alternative policy instruments like unwinding qe could be used to tighten monetary policy instead of using ior for this purpose.
As regards fed losses, they could reduce this a bit by buying more or just tips bonds, which would be less subject to losses.