The expectations fairy lives!

Many people are puzzled by my views on Fed purchases of assets. Long before the current pandemic, I was pleading with the Fed to ask permission of Congress to buy a wide range of assets in a crisis.

But I also suggested that I didn’t believe this would be necessary. Rather the purpose was to insure that a Fed “whatever it takes” approach to NGDP stabilization had credibility.

So was I the guy who favored having the Fed buy risky stocks and bonds, or was I the guy opposed to the Fed buying risky stocks or bonds? Perhaps this Bloomberg article will clarify the point I was trying to make:

The Federal Reserve’s planned foray into the $4 trillion exchange-traded fund market may be over before it ever began.

The U.S. central bank’s surprise announcement that it intended to purchase investment-grade ETFs as signs of a liquidity crunch mounted drove up demand from investors trying to front-run the central bank. . . .

For now, the Fed’s signal alone has been enough to loosen the cash crunch that had squeezed global bond markets amid spreading coronavirus fears. . . .

“Certainly anecdotally, finding buyers of the underlying bonds was becoming harder,” said David Mann, head of global ETF capital markets at the issuer. “The market has interpreted that announcement by the Fed as ‘OK, they’re going to be stepping in, so we can go back to normal,’ and if the threat of them acting was sufficient, then they don’t necessarily even have to act.” . . .

With the “healing process” already taking place, the Fed may not have to follow through with ETF purchases, according to Barings.

“Unless that high-yield market rolls over and comes under additional pressure, the Fed may never venture into buying high-yield ETFs,” said Scott Roth, a high-yield portfolio manager at Barings. “The signal that they could buy may be enough to achieve its desired effect.”

Just to be clear, the Fed still hasn’t adopted my suggestion of a “whatever it takes” approach to level targeting.  And they may ultimately need to buy these assets.  Nonetheless, this example certainly illustrates the basic idea that monetary policy is 99% about expectations, about credibility.  In a very limited way, the Fed won a battle without firing a shot.  This is sometimes called “the Chuck Norris effect.”

HT:  David Beckworth



32 Responses to “The expectations fairy lives!”

  1. Gravatar of Ironman Ironman
    16. April 2020 at 14:56

    In case you missed it in today’s news, but I suspect you haven’t:

    Fed Can Help Save Lives by Targeting GDP Growth: Dallas Fed Adviser

  2. Gravatar of Benjamin Cole Benjamin Cole
    16. April 2020 at 15:18

    This analysis strikes me as correct, as far as it goes.

    On another level, we see the stark limitations of purely monetary policy.

    There are now 20 million unemployed Americans, by federal and state ukase. To be sure, this is a case where the Federal Reserve is entirely not at fault. Evidently this number of unemployed will grow by about five million for every week of lockdowns

    In the first stage, these Americans must have the ability to buy food, pay rent and so on. Some people propose debt peonage, that is, lend these people the money to survive.

    And really, we propose to print money and give it to bondholders to make them whole, but not print money and give it to workers unemployed by government ukase?

    In the second stage these Americans must be re-employed as quickly as possible. Is that possible through purely monetary policy?

    If expectations are important, then what are the expectations of a future chained to heavier and heavier debts?

    Rather obviously, central banks need to be integrated into fiscal policy. As Stanley Fischer has indicated, old taboos are no longer useful, and are even self-destructive.

  3. Gravatar of Matthias Görgens Matthias Görgens
    16. April 2020 at 16:55

    Benjamin, you rightly point out that there are limits to monetary policy. But the real limit is that monetary policy can’t by itself fix supply problems like everyone being locked down.

    They are not giving money to bond holders willy nilly. They are buying the bonds at market prices.

    That first means that the sellers don’t particularly benefit from having dealt with the Fed. And second, it means that there’s more money floating around the economy to be earned by normal people

    Giving extra money to the needy might or might not be a good idea. But evaluating the wisdom of that action should be done completely separately from monetary policy.

    To the extent that any policy works as a stimulant or retardant to inflation the Fed will offset it. They have already shown their mettle in keeping inflation in check. They are currently also exploring their powers to proper spending up to a fuller extent than they have in the last few decades.

  4. Gravatar of Mike Sproul Mike Sproul
    16. April 2020 at 17:14


    Interesting that you favor the quantity theory while I favor the backing theory, but we both favor easy money policies. The backing theory asserts that new money, adequately backed, does not cause inflation. But a lack of money will cause a liquidity crisis. So, issuing new money will relieve a liquidity crunch without causing inflation, and restricting the money supply will cause a liquidity crunch without relieving inflation. Central banks should therefore follow a “never tight, always easy” approach.

  5. Gravatar of Benjamin Cole Benjamin Cole
    16. April 2020 at 18:45


    Yes, I think there is a role for conventional QE. I perhaps disagree with your analysis that the Fed is buying at “market prices.” One point of conventional QE is to prop up markets. Indeed, that is what proponents say about conventional QE; that the very expectation the Fed will buy helps prop up markets.

    I totally agree that supply side is important.

    But it also true your supply side will wither and die unless there is demand. When the US un-employs 50 million people (evidently the goal right now), and shuts down millions of businesses, I say you toss the rule books out and do whatever it takes to get them employed again.

    I hope President Trump puts his face on a $5,000 bill and sends one to every American, and gets the Supreme Court to give him authority to end lockdowns everywhere and unilaterally.

    If inflation runs around 5% for a few years, or NGDP runs at 10% up for a few years…a small price to pay to full employment.

    BTW, to anybody: When you have 50 million unemployed in the US…and the global economy shrinking, then what are your macroeconomic plans? I mean, besides prayer?

  6. Gravatar of Lorenzo from Oz Lorenzo from Oz
    16. April 2020 at 18:57

    Hence I try and explain to people that when the RBA changes interest rates it has meant something different from when other central banks change interest rates. Unless you understand that, monetary policy will remain a mystery.

    On backing theory, surely the point of fiat money is to trivialise the currency obligations of the central bank (or whoever is issuing bank notes).

  7. Gravatar of Benjamin Cole Benjamin Cole
    16. April 2020 at 20:02

    Add on note to Thaomas and Other Supply Siders:

    Okay, I have been a supply-sider my whole life.

    But…think about it. Globally, every industry is glutted with potential supply.

    There are more auto plants making cars than there are buyers, the same for smartphones, oil, crops, or any product or service you can name, with the exception of housing, and that is because construction in criminalized in places like Hong Kong and the West Coast of the US.

    I looked into potash a few years back…and the market became glutted.

    So supply and capital is overabundant…but aggregate demand is weak. So, what should policy be?

    You know, for 40 years some prominent economists predicted higher inflation and interest rates were pending, imminent, based on US fiscal and monetary policy. It never happened.

    So, we have been honking about the supply side for forty years too. But there are gluts of everything.

    Maybe we need to adjust.

  8. Gravatar of dtoh dtoh
    16. April 2020 at 23:24


    Yes and duh. You’ve been saying this for years, and why this is not glaringly obvious to everyone is a mystery. It’s unfortunate that most of your colleagues in the economics profession are not as smart as you.

  9. Gravatar of Steve Steve
    16. April 2020 at 23:45

    Sumner’s ‘bad’ blog is totes adorbs, but misses two points.

    First, the USA and world are already on the path toward herd immunity, thanks to the decisions of Winnie the Flu and DeBlasio too. It’s too late for containment.

    Second, the Fed can buy crap all day long, but the expectations channel is a chimera when it comes to risk assets. The Fed could buy oil futures, and make oil go up, but eventually people will realize the Fed can’t take delivery and everything crashes. What’s the Fed’s plan with junk bonds? They’ve triggered retail inflows, but is the Fed willing to take debtor-in-possession and subsidize restructurings? Or is the Fed going to screw taxpayers and sell to Blackrock at large losses while retail chasers gets reamed?

    This is why the Fed should just monetize the deficit and stay out of credit manipulation.

  10. Gravatar of Christian List Christian List
    17. April 2020 at 00:22


    Ah look at that, what a surprise: “China revises coronavirus death toll in Wuhan up by 50 percent.”

    The lack of corrections by the CCP has been one of the best indications that the death figures in Wuhan are wrong. The CCP now seems to realize this.

    In Madrid and Lombardy, the figures were corrected upwards several times retrospectively. The CCP has probably realized that it is not credible at all if they do not correct their own figures as well.

    What took the CCP so long? Obviously the surveillance state in Wuhan is not as developed as in Madrid and Lombardy, so of course such high percentages can be overlooked for a few months. Was this all by now, or what else is to come? Who is making a bet?

  11. Gravatar of Ray Lopez Ray Lopez
    17. April 2020 at 02:29

    Another brilliant article by Sumner, but he does not comment on whether the March Fed announcement about buying *stock* ETFs had an effect on the US stock market. Yes or no? I’d like to know what Sumner’s official views are on this.

    @Mike Sproul – what you propose is called “narrow banking” and will not stop inflation (“The backing theory asserts that new money, adequately backed, does not cause inflation”). That’s because assets fluctuate in value. Money backed by NYC real estate that has a radioactive bomb go off in it becomes worthless. Money backed by Houston real estate that unexpectedly has oil found in it becomes very valuable. And so on.

    @Christian List – countries revise their death toll higher all the time for C-19. Can you explain the French ‘spike’ in new cases? Also in Spain and Germ-any the last two days. And so on. I still think the Wuhan virus is Made in China, released accidentally by humans.

    @Ben Cole – again, what are your views on hyperinflation? Good, bad or indifferent? You’ve answered every question except this one, why?

  12. Gravatar of Ray Lopez Ray Lopez
    17. April 2020 at 03:32

    Other people also wondering whether the Fed is lifting US stocks by lifting the US economy (see below). I suspect Sumner will not answer my question above, since he’s too smart to fall into my trap (smile). You see, when you make predictions you must keep them vague, since reality has a way of nullifying anything you say. To say the Fed is or is not affecting stocks via QE would be a mistake by Sumner, but we’ll see what, if anything, he says. -RL

    Obviously, the mystery factor here is the psyche of investors. Do investors really believe the Federal Reserve and the federal government have done enough to shore up an economy that has effectively been turned off?

  13. Gravatar of Michael Rulle Michael Rulle
    17. April 2020 at 04:43

    Maybe I have fallen into that trap of circular thinking, but I don’t think so. I have always understood Scott’s “whatever It takes” paradigm. In fact, it helped me resolve one of my own conundrums about the Fed re: QE or whatever we want to call their buying securities in the open market. Before reading Scott, I believed the Fed holding a large amount of securities was a form of training the private sector to stand back and not make markets, i.e., moral hazard. I compared it unfavorably to 1907, when Morgan persuaded Banks and others to begin providing liquidity.

    But MM is about giving confidence, to otherwise fearful institutions. First, make sure the supply/demand for money is in “equilibrium”—-don’t know how that is done—-but it’s not done by being stubborn. Second, provide system wide liquidity——and once the markets know an infinite buyer is willing to do whatever it takes, they know there is opportunity in doing it themselves—-and no, the Fed does NOT WANT to do whatever it takes—-but has to be willing to do so.

    It’s unusual. Scott, and the great Monetarists, (i.e., Milton Friedman) in the past often speak or write in almost mechanistic terms——perhaps because there is so much understood complexity behind that which leads to “mechanistic” solutions.

    And no, the Fed has not said “whatever it takes”—have not said Nominal GDP targets are preferred, but have stuck with an explicit “symmetrical” price target—-and “That ain’t bad”—and have certainly walked up to the line of “whatever it takes”.

    In all matters of science—at the end of the day for we mere mortals, predictions that prove correct more times than random are what matters.

    And Scott has been doing it real time with his prescriptions—-and the legal trained Powell appears the most open.

    Still, perspectives of decision makers can change on a dime. Like a recent essay here said it’s output that matters——results.

  14. Gravatar of Ralph Musgrave Ralph Musgrave
    17. April 2020 at 05:10

    The whole expectations idea strikes me as weak. What proportion of households are going to run out and buy consumer durables or book an extra holiday just because the Fed announces it will “do whatever it takes”? My guess is less than 5%.

  15. Gravatar of Christian List Christian List
    17. April 2020 at 06:35

    countries revise their death toll higher all the time for C-19


    Western countries have done this, yes. The CCP has not done it so far, which was very suspicious because Wuhan was known to be the first cluster, well ahead of Lombardy, Madrid, NYC, so the corrections should have been well ahead of the other regions, but it was way behind.

    So how did the CCP “overlook” the dead and why did they turn up now? The German press writes: “The correction was made shortly after internal documents of the CCP have become known and have been distributed by the AP.” — What a coincidence, as soon as the documents got leaked, the CCP counts differently.

    The CCP says: It was in the early stage of the pandemic, you know how it was back then, total chaos, hospitals overwhelmed, people dying alone at home without ever being registered for corona. So in other words: It’s easy to miscount by 50%.

    The problem is, try to squeeze 50% more total deaths (!), or in absolute figures over 1000 deaths, into the curve of the early pandemic, see below. It’s a miracle.

    But well, I’m not here to solve the propaganda lies of the CCP. All their death curves are now messed up and no longer credible. They had drawn their first curves so beautifully, it is a drama.

    Nice exponential curves and perfectly linear with logarhythmic scales, and now they want to squeeze 1000 more deaths into the beginning? Wow. This is bold.

    Also very funny: The CCP says they just missed those sick people, see above, but even though the deaths were raised by over 1000, they only added 325 new corona cases.

  16. Gravatar of Aleksandar Aleksandar
    17. April 2020 at 06:39


    I am reading your blog posts about EMH in which you try to defend it because it suits well in your defense of market inflation expectations as guiding principle for monetary policy.

    I don’t know much about EMH but point I want to make is

    You don’t need to use EMH to defend Market Monetarism. There is one better irrefutable argument in favor of inflation expectation guided monetary policy.

    You, and others, stated it:

    Scott Sumner (2019):
    “But money is very different from other goods. Less demand for goods is contractionary for the economy and often leads to higher unemployment. Money is just the opposite. Less demand for money is expansionary for the economy, boosting employment. That’s because money is the other side of any transaction. People can reduce their holding of money only by purchasing goods, services, or financial assets.”

    Leland Yeager(1965):
    “The catch is this: while an excess supply of some things necessarily mean an excess demand for others, those other things may, unhappily, be money. If so, depression in some industries no longer entails boom in others”

    John Stuart Mill(1829):
    “It must, undoubtedly, be admitted that there cannot be an excess of all other commodities, and an excess of money at the same time.”

    As I understand, the same people who purchase TIPS are the ones who sustains from buying other goods. It is then tautology, unbreakable law, to say that when inflation expectations are falling that means that AD is falling in real time.

    Why are you using EMH to defend Market Monetarism when you have this superior argument?


  17. Gravatar of Spencer Hall Spencer Hall
    17. April 2020 at 06:41

    re: “Just to be clear, the Fed still hasn’t adopted my suggestion of a “whatever it takes” approach to level targeting”

    “For The First Time Ever, The Fed Will Monetize Double The Total Treasury Issuance”


  18. Gravatar of Benjamin Cole Benjamin Cole
    17. April 2020 at 07:24

    Dr. Ray Lopez:

    I think hyperinflation is a poor outcome. I realize you do not hang on my every word, but many times I have written that inflation below 5% is not that much of a concern.

    Forgotten today is there was a time when even monetary economist did not obsess with minute rates of inflation. In 1992 Milton Friedman chastised the Federal Reserve for being too tight, when the CPI was registering an inflation rate of about 3%.

    On the other hand, when the Federal Reserve deployed its 2% IT after 2008, the most anemic recovery of all time happened.

    So why do American macroeconomists slavishly genuflect to the 2% it totem? I don’t know,

  19. Gravatar of Todd Ramsey Todd Ramsey
    17. April 2020 at 07:53

    Prof Sumner-

    Should the Federal Reserve buy TIPS?

    Would buying TIPS add excessive noise to a useful data source?

  20. Gravatar of myb6 myb6
    17. April 2020 at 08:41

    Fed seems rather quiet on the expectations front, even on plain old inflation targeting. Working out inflation expectations for individual years (not cumulative) thru 2025 using WSJ Treasury/TIPS yields, no year is even sniffing at 1.5% (if I did my math correctly). What gives?

  21. Gravatar of myb6 myb6
    17. April 2020 at 08:49

    The Fed has stated “whatever it takes” to “shore up the economy”. Do they need to more specific? To “get spending back on path”? “Unemployment back down”? “Make up for the lost time”?

  22. Gravatar of ssumner ssumner
    17. April 2020 at 09:54

    Thanks Ironman. He’s been along time advocate of NGDPLT.

    Mike, I don’t see liquidity crises as the main problem. I don’t recall any liquidity crisis in 1982, but tight money caused NGDP growth to plunge. Because of sticky wages, slower NGDP growth raises unemployment, even if there is no financial stress.

    dtoh, Thanks, but plenty of them are much smarter than me.

    Steve, NYC will get a modest amount of herd immunity (not total), but out here in Orange County we will not. We’ve hardly been touched, and we are almost half way though the first wave.

    I agree that the Fed should focus on increasing the money supply, not providing credit.

    Christian, Thanks for pointing that out; that revision is almost precisely what I’ve been saying. Will you finally now accept that I was right about China? If they were covering things up as you claimed, then why do this revision?

    You said:

    “Also very funny: The CCP says they just missed those sick people, see above, but even though the deaths were raised by over 1000, they only added 325 new corona cases.”

    Yes, it looks funny if you have no idea what’s going on in the world. We knew back in January that some Chinese coronavirus deaths were called “pneumonia” because the coronavirus patient actually did catch pneumonia and died of pneumonia. I pointed this out in earlier posts, but I wouldn’t have expected you to pay attention. So yeah, the world looks “stupid” when you don’t know what you are talking about.

    Ray, The Fed is buying stocks?

    Ralph, Consumption is not the goal of monetary policy, higher NGDP is the goal.

    Aleksander, You asked:

    “It is then tautology, unbreakable law, to say that when inflation expectations are falling that means that AD is falling in real time.”

    No, It’s not a tautology. It’s a theory, and not a good one.

    Spencer, That headline is not very interesting.

    Todd, Yes, TIPS are fine, in moderation.

    myb6, Yes, they need to be more specific.

  23. Gravatar of ssumner ssumner
    17. April 2020 at 10:02

    Christian, Here you go:

  24. Gravatar of Christian List Christian List
    17. April 2020 at 10:58


    you only read what you want to read and what fits into your mindset. It’s astonishing. If the CCP now simply adds over a thousand deaths at the beginning, then the whole curve no longer fits. It’s just ridiculous.

    So the wild assumptions of some of the studies you linked to are also wrong, as I suspected.

    Most curves assume deaths in the low double digits until almost the end of January. These are steeply rising curves, just add over a thousand deaths in January. I can only encourage that, the earlier the better. January you say? Great, go on.

    We knew back in January that some Chinese coronavirus deaths were called “pneumonia” because the coronavirus patient actually did catch pneumonia and died of pneumonia.

    You’re not good at writing about medical details, Scott. How do you think people die from corona? Do aliens abduct them? If you die from corona, it’s usually from lung inflammation, the medical term for it is pneumonia.

    If the CCP really did what you’re saying, it’d be like saying, “people who have heart attacks did not die of heart attacks because they had a heart attack.” It doesn’t make any sense and creates new confidence in the ever surprising CCP system.

    Bonus question: If you are right, what are the causes of death of the people who were previously counted as corona deaths by the CCP? Apparently not pneumonia. Is this where the alien abduction comes in?

    Now I know why Wuhan has so few deaths: pneumonia deaths are not corona deaths. Scott, you have solved the mystery!

    If they were covering things up as you claimed, then why do this revision?

    I can’t confirm this yet, but Western media writes that the CCP only corrected what they had to correct, because the data has been leaked to the Western press already. I haven’t read the reports myself, but I don’t see why the Western media should lie about it either. I trust them more than the CCP, that much is certain.

  25. Gravatar of Spencer Hall Spencer Hall
    17. April 2020 at 11:59

    “Fed Cuts Pace Of Treasury QE By 50% To “Only” $15 Billion Per Day, Yields Spike”


    How about this one that will tank the markets?

  26. Gravatar of Effem Effem
    17. April 2020 at 12:02

    And the Fed inequality-machine just keeps on chugging along. One day it’ll be the Fed political-volatility machine but we’re clearly not there yet.

  27. Gravatar of Monetary stimulus is not aimed at consumption – The Rabbit Hole Monetary stimulus is not aimed at consumption - The Rabbit Hole
    18. April 2020 at 00:48

    […] commenter recently […]

  28. Gravatar of ssumner ssumner
    18. April 2020 at 09:08

    Christian, Perhaps you should study up on “comorbidity”. By the way, the exact same data problems exist in lots of countries, which are also raising their estimated death totals, often by amounts even larger than in China. Take a deep breath.

    You could say my dad died of “emphysema” or that he died of “flu”, and either statement is accurate. Not everything is a conspiracy.

  29. Gravatar of Aleksandar Aleksandar
    19. April 2020 at 02:50


    Thanks for answer.

    You said:
    “No, It’s not a tautology. It’s a theory, and not a good one.”

    I though that is your theory and also theory of Market Monetarism?
    Can you explain a bit more why it is not correct or provide me some links about it?

    See Milton Friedman’s video. He says that FED should target Inflation expectations. I am going further and say – you don’t need FED for that, it could be done by an automated computer/machine.

  30. Gravatar of ssumner ssumner
    19. April 2020 at 08:52

    Aleksander, Inflation expectations might be falling because AS is rising faster than usual. That doesn’t happen very often, but the alternative is not a “tautology.

  31. Gravatar of Aleksandar Aleksandar
    19. April 2020 at 09:55


    You are right. But if we substitute inflation expectation with NGDP expectations it will work.

    The point I want to make is that inflation expectations can serve as proxy for money demand for now until we find something better.

    But this still stays: you either buy money or other goods. Walras’s Law is a tautology. I don’t know what is the problem here?

    This logical arguments is still better than EMH, because you can not refute it. The catch is only to find reliable indicator of the demand for money.


  32. Gravatar of Mark R Mark R
    22. April 2020 at 08:53


    I’ve been re-reading your blog again – so many valuable lessons that need to be imbued numerous times to get right!

    Back in 2012 / 2014 you were having persistent arguments with the Austrians about whether QE caused inequality through Cantillion effects (i.e. the real price impacts and subsequent decisions made through the economy). Clearly one of the arguments raging now is about whether QE of private assets is a bailout of capital owners at the expense of employees. Your old argument here seemed to be that raising reserves in place of government spending (eg paying government employees by increasing reserves vs buying bonds which funded the spending) is a wash – do you think that is still true here? i.e. targeting specific sectors / segments will surely have different distributional effects (through the Cantillion effect) depending on who / how the reserves are targeted at the private sector? If nothing else, it has created a moral hazard for shareholders and managers when faced with paying themselves vs. creating a resilient business. How should that risk transfer accounted for?

    Second Question: One thing that I have been trying to think about is the dynamics of wealth accumulation and how QE affects that (similar to the difference between the static vs dynamic tax arguments – i.e. comparing tax brackets at static points over time is useless unless you also analyse the dynamics through the tax brackets over time). Over time, the youth work to buy up assets which are owned by the old. If asset prices rise but incomes do not (certainly not to the same extent) then isn’t there an implicit wealth transfer to the old? This transfer may be caused by the CB not following through on their promise for inflation forcefully enough, but to a certain extent that doesn’t matter – if it is half measures then it is still detrimental to the youth (and probably exacerbates inequality in a multi generation framework)?


Leave a Reply