Why Didn’t QE Cause High Inflation?

Larry Kudlow has declared the deficit hawks (the inflationistas) the losers of the long raging inflation debate (see here for details). After almost 5 years of QE and thousands of pundits shrieking over high inflation the debate appears to have been convincingly won by the inflation doves (those who thought inflation would be low despite QE).  But Larry Kudlow asks an important question – why didn’t QE cause high inflation?   Let’s take a look.

First of all, I think QE did cause some inflation.  Contrary to popular mythology, I don’t say QE does nothing.  It’s more that I just think it’s inadequate.  And when we analyze the current environment and impact of QE we have to consider the counter-factual world without having had any QE.   If the Fed hadn’t stepped in in 2008 to bolster the banking system we likely would have had a depression in the USA.  The payments system would have seized up for months and maybe even all of 2009 and thousands of businesses would have gone under as bank defaults rippled through the economy.  This would have been hugely deflationary.

When the Fed stepped in to make a market in MBS in 2009 they were basically creating capital gains to the tune of billions of dollars for the banking system by announcing that all the GSE assets were worth 100 cents on the dollar.  And this was in addition to their lending facilities and the other bailout programs.  So QE1 had a big impact on the economy in 2009 and 2010.

But what about QE2 and the subsequent programs?  Why haven’t they kept the inflation train going up and up?  Here, I think it’s incredibly important to separate monetary policy from fiscal policy.  The tendency is for people to look at the Fed’s purchases of bonds as financing the deficit.  I think that skews the reality and implies that the Fed is needed to buy the bonds (as if there would be no other buyers without QE).  We know this is right because QE2 already proved it in real-time when big bond investors like Bill Gross said “who will buy the bonds” when QE2 ended and people like me said “you will buy them!”  The “monetization” crowd was wrong as yields tanked after QE2.  There was no shortage of buyers for Treasury bonds at all.  So let’s just look at QE for what it is – open market operations and not necessarily a part of fiscal policy.

Now, I don’t think it should be controversial to say that spending is a function of income relative to desired saving.  Inflation is an extension of spending.  And if producers have pricing power due to high aggregate demand or aggregate supply shortages then prices will generally rise (I’m oversimplifying, but for the purposes of this discussion that’s sufficient).  So you generally need the spending if you’re ever going to have the inflation.  It’s that old demand thing.  If you’ve ever run a business you know that revenues and pricing power don’t exist without customers walking in the door.

The problem with QE is that it doesn’t have a transmission mechanism to substantially increase aggregate demand.  When the Fed buys bonds from a bank they simply swap reserves for t-bonds.  The bank has the same net worth (roughly, depending on any capital gains and as mentioned previously QE1, 2 and 3 have had diminishing returns here) and the reserves sit in the interbank market (and no, they don’t get “lent out”, that’s not how banking works).  The bank might feel inclined to shift its portfolio holdings and replace lost T-bond income so it might go buy stocks or bonds of other types, but this doesn’t guarantee sustainable capital gains because there has been nothing directly attached to this balance sheet change that necessarily justifies an increase in share prices.

When QE is done via a non-bank the non-bank gets deposits, sells the t-bond to the bank and the bank does their reserve for t-bond swap with the Fed.  Again, there’s no change in private sector net worth and no change in incomes.  So, if we go back to our original understanding of inflation (that spending and ultimately inflation, is a function of incomes relative to desired saving) then it becomes rather obvious why QE hasn’t caused high inflation.  It hasn’t increased incomes.  And it hasn’t increased savings (except mainly for wealthy Americans who own stocks and bonds).

So, what it all really comes down to is this:

1)  How much does QE impact rates which can influence lending and investment?   I’d argue not that given that QE should have bolstered inflation, but it has actually fallen which has resulted in bond markets bidding prices higher (and yields lower).  

2)  How impactful is the “wealth effect” and portfolio rebalancing effect?   This is the meat of QE.  It has powerful psychological effects, but I think the evidence of sustainable capital gains from QE is weak.  For instance, why hasn’t QE in Europe supported the peripheral stock markets?

3)  How much does QE help shore up bank balance sheets?   QE1 probably helped a lot.  The subsequent programs probably haven’t helped nearly as much.  

4)  How much more does the private sector spend when they swap t-bonds for cash?   If the wealth effect is substantial then those in the upper class probably spend a good deal more as the value of their cash and bonds increase.  Whether this can be sustained or helps the broad economy is up for debate still.  

5)  Does QE drive down the dollar relative to other currencies leading to more competitive trade?  The USD basket is essentially flat since QE started in 2008 so the answer is definitely no.  

That’s all a bit oversimplified and it’s not my intention to try to prove that QE does nothing, but if you understood all of this 5 years ago you steered well clear of any hyperinflation or even high inflation predictions.  In other words, it looks like QE isn’t everything it’s trumped up to be….

See my QE primer here for a more details look at this explanation.  


Got a comment or question about this post? Feel free to use the Ask Cullen section, leave a comment in the forum or send me a message on Twitter.
Cullen Roche

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services. He is also the author of Pragmatic Capitalism: What Every Investor Needs to Understand About Money and Finance and Understanding the Modern Monetary System.

More Posts - Website

Follow Me:

  • Smob

    Thank you, clear and to the point. My question is one that many have asked which is why is the stock market flying high? My understanding is interest payments savings boosted corporate profits, could you elaborate on that and if it has any link to QE and how long will corporation be able to enjoy those savings?

  • http://wudicapital.com D

    Correct me if I’m wrong Cullen, but QE in general decreases the amount of whatever the central bank buys from the private sector, say treasuries. The decrease in supply thus increases its value and lowers rates. Thus QE is not even special as that’s what central banks have done since day 1. Only difference is QE targets the intermediate/long term yields since short term yields are already near 0%. That’s pretty much why QE hasn’t caused any inflation. No matter how low rates are, if households and businesses don’t borrow, there is no money creation and thus no inflation.

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Cullen, I formulated a question for rwperu34 here on a similar topic:


    But let me do it again fresh here. Let’s make some simplifying assumptions… namely that QE is no longer like QE1: i.e. it’s no longer used for explicitly shoring up bank BSs by buying MBS for face value (when they’re not worth face value), etc. Also, let’s assume:

    1. IOR = 0
    2. No reserve requirements
    3. No cash (an all electronic money system)
    4. Just one commercial bank

    Now let’s say we take your typical MMist (Scott Sumner for example) and put him in charge of the central bank in this world. The MMists claim (when pressed) that cash is NOT necessary for MM ideas to work. So now how does an MMist in this world go about NGDPLT? They still announce the central bank’s intentions to target NGDPLT… but how do they back it up? Buying Tsy debt and MBS isn’t going to do much is it? Nobody in this world really has any use for central bank deposits (reserves), so what’s that going to accomplish? How is the expected supply vs demand for “base money” (reserves) going to be of any importance? Why would there be any demand for “base money” at all? Do you think the MMists are sincere when they claim that cash is not required for their plans to work?

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Also I should have stated that this world is in recession and at the ZLB… like ours. So that means hitting a 5% NGDPLT means more inflation is needed.

  • http://internationalmonetary.wordpress.com LXDR1F7

    “5) Does QE drive down the dollar relative to other currencies leading to more competitive trade? The USD basket is essentially flat since QE started in 2008 so the answer is definitely no. ”

    QE IMO does drive down the dollar but it is only one factor out of several affecting the USD.

  • http://internationalmonetary.wordpress.com LXDR1F7

    There is money creation when the fed buys treasuries from non banks Im pretty sure. New deposit is created for non banks in exchange for treasuries.

    Indirectly money creation is happening because a lower supply of bonds from QE is allowing the government to issue more treasuries which they exchange for deposits. I maybe wrong if the treasury receives reserves when they issue treasuries.

  • Stephen

    We already know that is it about 80% of QE has been simply held in excess reserves which translates to zilch for inflation. We have confirmation that the transmission is what cents on the buck? Don’t know the number the than it low. So we really know that the the relatively small amount of QE that found it’s way into the economy was primarily an offset for the banking system. That is ,it mainly allowed them restore their balance sheets in exchange for the significant write downs of bad debt etc etc. Why should we wonder that inflation in price consumables was very low? Why would it be anythingelse when the effects on incomes etc has been relatively paltry. Plenty of inflation in asset prices of course, but we know what they mean to central banks when they consider inflation…nothing much.
    Central bank monetary policy seems to work much better on assets than incomes and an higher order of inflation really needs the latter which has continued to struggle.

  • http://expectingreturns.com/ CF

    “spending is a function of income relative to desired saving.” – And a key factor in that statement is demand for credit. If people are not borrowing more from banks to finance spending (because they are still paying down existing debt or are just not confident in their employment situation), inflationary pressure has to be lower.

    Furthermore, if banks are lending less because of a host of regulations it makes sense for banks to park an additional liquidity in t-bills and stocks rather than loans.

    The (hardcore) inflationistas would argue that a jump in credit growth can happen unexpectedly and when it does it will not be controlled. Do you think this is a risk?

  • Nils

    It did lead to higher prices, you’re just to ignorant. I raised prices for my gold newsletter thrice.

  • jim

    In Los Angeles and Orange County, in cities like Manhattan Beach, Newport Beach, Corona Del Mar, … bidding wars are breaking out for many homes. Many with multiple all cash offers. Prices are exploding. I have been involved with real estate in this area for 25 years. I have never seen anything like this. This real estate market is on fire. Buyers that have a good down payment, and are preapproved for a mortgage are out of luck. Their offers are often rejected, and they are getting angry. This is the beginning of something very big … like rapid inflation. It is hard to tell how high prices might rise. Scary stuff.

  • Geoff

    Excellent QE Q&A. I think #2 and #5 might be related in the sense that any psychological effect would also probably affect the USD. But if so, it would probably only be temporary as well.

  • jswede

    “When the Fed stepped in to make a market in MBS in 2009 they were basically creating capital gains to the tune of billions of dollars for the banking system by announcing that all the GSE assets were worth 100 cents on the dollar.”

    MBS was trading well over par before the Fed started purchasing, and have been ever since. MBS prices, if anything, and much like USTs, dropped on Fed buying.

    what do you mean?

  • Anon

    Two words: housing recovery …

    Housing is such a huge segment of the economy, and it has barely bounced back – let alone has it caught up with demographic pressures.

  • Anon

    Hopefully people sold their gold before gold price tanked?

  • jaymaster

    I’ve been toying with a theory that QE3 might be stimulative in another way. By guaranteeing a market for MBS, it encourages banks to offer mortgages under conditions where they might not normally want to do so.

    I’m not necessarily talking about them offering “risky” mortgages (i.e, ones with a high potential rate of default), but instead I’m thinking about interest rate risks. Those MBS’s that are backed by 3-4 percent mortgages are going to turn into stinkers if/when rates rise.

    But as long as the Fed is willing to buy them, the banks can be content to make some profits on the mortgage processing fees, and then immediately dump those mortgages on the Fed (or hold them now, and dump them if rates start rising while QE3 is still in effect. So the Fed is essentially taking on the rate risk instead of the banks.

  • http://orcamgroup.com Cullen Roche

    My guess – how about the govt spending a trillion dollars every year for 5 years?


  • http://orcamgroup.com Cullen Roche

    Right. With the deficit QE is about asset neutral though since new t-bonds are issued. It’s duration negative though.


  • http://orcamgroup.com Cullen Roche

    It’s all about expectations (well, mostly) in the MM world. The Fed doesnt have to prove it has a bazooka. It just has to claim it does. The old Chuck Norris thing. I prefer Bruce Lee personally. “Saying is not enough. You must do.”

  • http://orcamgroup.com Cullen Roche

    Jump in borrowing would most likely occur in an investment boom. So we’d having a manufacturing and real estate bubble or another tech bubble. That’d be good until it goes bad….

  • http://orcamgroup.com Cullen Roche

    Shadow Stats didnt get the price raise memo!!!!


  • http://orcamgroup.com Cullen Roche

    Same thing going on in San Diego.

  • Geoff

    I managed to get off some of my gold bullion (I admit I was once gold bug) near the high in my own version of QE (I swapped gold for cash :)

    But I still have some. I am seriously considering taking advantage of the recent rally to unload the rest.

  • bubbleRefuge

    Same thing in bay area and south florida.

  • http://www.aymankhlifat.com/ ayman khlifat

    Thank you , great post ! , keep it up .

  • http://howfiatdies.blogspot.com/ Vincent Cate

    The Fed made new money but has only let 0.25% interest on most of that money actually leave the building. So it mostly is not causing inflation yet. Bernanke’s trick of paying interest on “excess reserves” that is more than short term bonds has worked so far to keep the money off the street. As interest rates go up then the banks will find better uses for the money and will take it out of the Fed. Then we will get the inflation.

    Cullen, you got me all excited when you said you were going to post a reply to my Hyperinflation FAQ but I am still waiting!


  • http://orcamgroup.com Cullen Roche

    No time yet. Sorry. I am juggling lots of different things right now….

  • http://howfiatdies.blogspot.com/ Vincent Cate

    Ok, just don’t forget. :-)

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Vincent, let me try out this argument on you. First humor me and allow me to make some simplifying assumptions on our monetary system (I’ll justify these later):

    1. No reserve requirements
    2. No physical cash (paper reserve notes or coins)
    3. Only a single commercial bank

    I claim that these three assumptions should not effect the macro picture much, but that they go a long way to simplify the accounting. We can add to this list

    4. No IOR

    if you wish. Now which sellers can our one commercial bank buy from? I claim they fall into two classes, each of which requires a different means of making a purchase:

    (i) Fed deposit accepting sellers. This group itself is divided into the Fed itself and all other Fed deposit holders (e.g. Tsy). Strictly speaking the Fed doesn’t hold deposits as assets, only as liabilities, so it’s different in that regard.

    (ii) Non-Fed deposit accepting sellers: this includes everybody else (the non-bank private sector). What does this group accept for payment? Bank deposits and nothing else. They are not legally allowed to hold Fed deposits!

    Let me state up front another simplifying concept: I claim that when a bank makes a loan, it essentially buys a loan agreement from the borrower. This simplifies the bank’s activities: it’s activity consists solely of purchasing and selling items.

    What can the bank purchase using method (i)? Well what are those sellers selling? Pretty much they are selling one thing: Tsy debt. That’s it! That’s what our bank can buy with it’s reserves. If the Fed were selling (which it’s not), it could by some MBS too. Not very exciting.

    Now what about method (ii)? With this method it can buy pretty much anything under the sun, including 2nd hand Tsy debt, loans, donuts, real-estate, office supplies, stocks, etc. How does it make these kinds of purchases? It credits the seller’s bank deposit. It has to abide by capital constraints and stay solvent… but that’s it.

    Now explain to me how the bank finds “better uses for the money and will take it out of the Fed.” The money it holds as a Fed deposit can ONLY go to category (i) sellers, and thus there’s limited wares to buy from them! How does this cause inflation? The items it buys from category (ii) sellers are items in the real economy including every conceivable financial and non-financial asset. These purchases are NOT affected by the balance in its Fed deposit, since it creates the money ex-nihilo to make these purchases with or without any Fed deposits (i.e. it credits the seller’s bank deposit).

    OK, now for the justifications for my three assumptions. This will be quick.

    1. No reserve requirements (RRs)… assume we adopt the Canadian system. There’s lots of evidence that banks are not reserve constrained.

    3. A single commercial bank: Aggregate all the banks together and you have exactly the same picture from the macro perspective.

    2. No cash: I claim that cash is purely a function of the desire and ability (i.e. having a deposit to withdraw cash from in the first place) of the non-bank private sector to hold it. This desire is not influenced in the least by the existence of excess reserves (ERs). The Fed provides for the public’s need for cash (working in conjunction w/ the banks) but if there was no cash, we’d make due just fine.


  • http://brown-blog-5.blogspot.com/ Tom Brown

    Yes, that’s a good point. But what does that say about Sumner’s precious EMH when this expectation is based on irrationality? Doesn’t somebody out there *get it* … that the emperor has no clothes… and then the whole house of cards falls apart? Isn’t that what the EMH would predict?

    I’m going to guess that if they are serious about cash not being integral to the process, then in the scenario I laid out here, the only real way to actually do something to back up a publicly targeted NGDP level, is to let the Fed make “non-traditional” purchases. If that’s what happens, then (as you’ve pointed out) we’ve left the realm of monetary policy and entered the realm of fiscal policy.

  • http://howfiatdies.blogspot.com/ Vincent Cate

    If you make a model where you can not take money out of the Fed you should not be amazed if you can not take money out of the Fed.

    But in the real world banks can send an armored truck and take out a truck full of cash if they have “excess reserves”. They can then loan this cash to someone who pays them more, for whaterver purpose they want to borrow it for. Perhaps they are speculating that the price of gold is going to go up or taking advantage of the low rates to replace most of their workers with robots. Once they have the truck full of cash there is no limit to what projects the bank might loan it for.

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Ah!… that’s interesting. So you’re in the camp that thinks that cash is important. The MMists (like Scott Sumner, Nick Rowe, and David Glasner) seem to have a foot in this camp too.. although when pressed they back away and claim that cash is not important to do NGDPLT, to raise inflation, or for MM ideas to work.

    It sounds like you’re NOT willing to back away from that idea, true?

    Why do you think that cash is anything other than just a convenience? Do you think that if cash were eliminated that would squelch the link between QE and inflation then?

    I don’t have the data, but I’d be surprised if most loans were cash loans. I personally have never taken a cash advance in my life. I know I didn’t see a cent of cash when I bought my three properties on credit… nor do I see it when I make 99% of my purchases (either with my bank card or credit card). That’s just my personal experience… so it’s hard to extrapolate, but I admit I’d be shocked if the existence of cash is important.

    Aren’t most house and car loans and credit card purchases made w/o EVER touching any cash?

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Looking again at your examples “Perhaps they are speculating that the price of gold is going to go up or taking advantage of the low rates to replace most of their workers with robots.”

    … and you imagine this kind of thing would be primarily done with cash? If I were speculating on the price of gold, I’d go into my IRA brokerage account, and buy GLD. I would never see a cent of cash or the glitter of any gold. Even if I borrowed money to do it (i.e. I could take a loan from my 401k… as I’ve done on several occasions). Never ever though has this transaction been accomplished using cash. Even my Spanish speaking tenants (who probably don’t have bank accounts) pay me with money orders. I never see cash from them.. (although they probably use it quite a bit I suppose).

    My company sells robots. Nobody has ever paid us in cash! If they did, I’m sure we’d deposit it again ASAP (thus pushing the reserve levels right back up again!).

    I don’t see the mechanism by which more loans are made in cash because excess reserves exist (those excess reserves are MOSTLY in the form of electronic Fed deposits, BTW). What could that mechanism possibly be? I don’t know about you, but I feel no incentive to request loans in cash now that there’s lots of extra electronic reserves on bank balance sheets. What’s the connection?

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Also it’s extremely rare that I’ve ever taken out a loan for the sole purpose of having some extra funds available to play with… I did something like that once when I took a 401k loan (the bulk of which was used for another purpose). But the vast majority of loans (mortgage & credit card purchases) are for the exact amount needed to purchase the item I’m taking the loan for. There’s never any left over tempting me to go down to the bank and make a withdrawal in cash. Even if there were, I’d wager cash withdrawals would be a rarity.

    Maybe I’m a weirdo here, but isn’t that how most people operate? Who takes a loan out just so they can have a pool of money to withdraw in cash? Who’d give them such a loan?

  • Sailwind

    Washington DC and the suburbs (MD and VA) are experiencing the same squeeze of housing inventory. Demand is far higher than offer, hence the bidding wars, homes sold above listing, and multiple offers during open houses. Bubble all over again!

  • Sam Brown

    Very well said Mr Roche. You said what I have been saying since QE first began. The destruction of the American Middle Class, the Unions, and outsourced good paying jobs have repercussions. The repercussion is a much lower demand curve when looking at a simple economic supply/demand chart. Our worst inflation years, those of the late 1970’s and early 1980’s, corresponded with the baby boom generation coming of age. The “pig” in the python brought with it demand for education, for housing, and for other material goods. Rising personal demand coupled with unstable commodity prices, cartel creation, and global political instability led to the high inflation of the time. Now we are causing inflation in health care, medications, and soon, retirement homes and cemetaries. Keep up the reasonable analyses.

  • Sam Brown

    Who has a HELOC?

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Good point… but even there, do you draw out much in cash? I had one for a while… but I can’t remember a time when I ever paid a contractor or Home Depot with cash (yes shockingly I used my HELOC on my home).

  • Edward

    yen carry.

  • http://howfiatdies.blogspot.com/ Vincent Cate

    I am not saying that physical cash is important. I don’t think it matters if it is on a computer or paper money. But sometimes people get confused when they think about accounts on computers and it is easier to understand if you think about paper money.

  • http://brown-blog-5.blogspot.com/ Tom Brown

    OK, fair enough, so does that mean you don’t object to my simplification in eliminating it then? I guess my point is that cash is the only way that I can see that excess reserves can really leave the banking system. I don’t think that happens because I think circulation levels of cash have more to do with the public’s desire for convenience that anything else. The other two assumptions I make (to simplify the accounting of our system) don’t really have any possible impact on reserves leaking out. Reserves can really only go three places:


    I can make an argument why none of those three represents a path to inflation:

    1. Doesn’t go into the real economy. Basically the Tsy is the only important one in this category because everything else is just getting passed around between banks. If the Tsy runs a surplus, that’s a net flow OUT of the banks, but that’s not inflationary. Plus that’s not happening.

    2. Cash we covered.

    3. Back to the Fed is also not inflationary, and it’s not happening either.

    I don’t see why lowering IOR to 0 would make a significant change to any of those three paths.

    There’s really only those three possibilities for the flow of reserves OUT of the banking system.

  • http://brown-blog-5.blogspot.com/ Tom Brown

    Vincent, you write “The Fed made new money but has only let 0.25% interest on most of that money actually leave the building.”

    The Fed doesn’t pay any IOR on money that “leaves the building.” They only pay on bank Fed deposits, which do NOT leave the building.

  • Greg H.

    Ok QE is now buying back government bonds at the rate of 85 billion or so a month. So since a lot of foreign governments own a lot of our bonds, how much of the 85 billion goes in their pockets verses US banks , companies, or individuals? Can the Fed pick and chose which bonds to buy back or have we been just sending billions to China every month?