QE chatter

3 views
Skip to first unread message

Bijou Smith

unread,
Sep 29, 2025, 9:15:22 AM (6 days ago) Sep 29
to Modern Monetary Theory
Hi All,
This one's esp. for Ryan and others in the group who know some of the nuts & bolts of asset price dynamics. I passed it by Warren for starters, and he said "QE is not a driver of inequality per se."  

But thought I'd give you all a chance to help me respond in more depth. 

Jim Byrne from mmt101 wrote this, on a episode he put out with a bond trader, Bill Bain: 

Bill is sincere in his beliefs - he is not disingenuous and as you will discover as we chat - he’s a nice guy. And he knows a lot about the finance sector.
One other thing I will mention- so as not to confuse those who are MMTers who know a bit about Quantitative Easing - which we chat briefly about: is that it sounds as if I agree that QE is about the government dumping lots of money into the economy. I don’t agree.
What I’m saying is that it is from the additional liquidity that banks invest in things like houses and other assets (such as the stock market) - which drive up their value. In other words one of the things we can say about QE is that it is a driver of inequality.

Jim invited me into his substack writers discussion group, so I was hesitant to write a reply on this off-the-cuff. But what I was going to write in reply was this:
---

Anyone care to comment on Jim's recent chat with this guy Bill Bain?

Asking since I don't think the highlighted paragraph there is MMT aware. Banks do not need reserve liquidity or "additional liquidity" to lend, the loan creates the deposit. They only need credit-worthy customers. The spread rate the bank charges always covers their anticipated marginal costs. The reserves are always dealt with by the reserve accountants independent of the loans desks. If this does not apply to the UK/Scot banking system then my mistake (maybe you guys are promising to deliver gold nuggets"?) If anything, QE is a loss of interest income, so is all things considered deflationary biased or at least asset price neutral (if you want to say there is any impact at all, from investors moving scorepoints from savings accounts to check accounts).

Plus, I am not sure what he is implying about "bank investment". Issuing loans for the buyers? Or the banks using the extra cash from interest payment on excess reserves to go and purchase assets on their own books? But unless there is aggressive bidding I don't see this pushing up prices.

I'd ask Jim: How exactly do you think QE effects asset prices? It doesn't to first order imho, and cannot long run (since it is not a flow variable). There might be a "zeroeth order" psychological effect, but that can only ever be short term and transient. I'd even say the bonds are what support higher asset prices, not the reserves. The interest income from bonds are massive amounts, which disappear with QE, modulo interest paid on reserve balances for the bankers, which is just defining a component of the base inflation rate.

I do not think QE effects the housing market for example, nor "asset prices" generally. That's a whole different story, a monetarist's story as a fiction, and a real-estate industry bundle of misaligned policy incentives, plus some corruption and fraud and whatnot as a fact. It is worth digging in to what truly causes asset bubbles and general relative price adjustments in assets, boosting them relative to wages, since there is so much of a tsunami of disinformation from the other side.

Bain talked about government bond investment being about the risk of repayment. He claims, "We demand a higher rate if we perceive more risk..." But is that really the case, since the government can always make the payment, they just might not be willing to, but except in extreme cases (Russian ruble in the 90's?) do they ever default on bonds? From my understanding the risk to a bond trader is all relative value: is the bond rate going to stay ahead of their index of concern, like CPI or whatever? That's the risk. Could the currency have been better invested in other equities or securities, not about the risk of a default.

Bain also claimed (rather forcefully)): "The bond market is the most important indicator of what is going on in an economy." Unemployment and inequality are by far the more important indicators, and it's not even close. The bond market could be Harry Potter vamoosed and no one should care at all, the bond traders can get other jobs with their skills.

---

Is that a fair comment for Jim?

I'm also asking this group since I don't like seeing things going out under the "MMT" banner if they are not actually MMT consistent. I would not expect any of you to sit and listen to the whole thing, but if you wanted to give me you thoughts then here's the link:

https://mmt101.substack.com/p/bill-blain-bond-trader-ceo-of-wind?publication_id=2768084&post_id=174692110

I have not annoyed Jim too much yet, so have a window to respond with some decent insights.  

Cheers,
-Bijou

Jay Mills

unread,
Sep 29, 2025, 9:36:13 AM (6 days ago) Sep 29
to Bijou Smith, Modern Monetary Theory
Bijou,

QE doesn’t pour money into the economy or give banks new funds to lend. It swaps bonds for reserves, changing the composition of financial assets but not their net quantity.  Is this correct?

Banks don’t need reserves to lend—loans create deposits. Reserves settle payments afterwards. Lending depends on capital and credit demand, not “extra liquidity.”

QE is just the Fed swapping higher-yield bonds for lower-yield reserves, so it’s not new money for lending. While the buying flow can lower long rates and mortgage spreads for a while, the effect fades when the purchases stop, and at the same time bondholders lose interest income. That means QE isn’t a one-way pump into asset prices or inequality—it’s mixed, with push-up and pull-down effects. A currency-issuing government can always make payments, so yields mostly reflect inflation and relative returns, not default. And bond yields are trader signals, not measures of jobs, wages, or inequality, which are what actually matter. So my question back is: through which concrete channel do you think QE drives inequality—temporary portfolio shifts, mortgage refinancing, or something else—given that none of these is permanent or especially large?

My questions:
  1. Which specific channel do you see as driving asset moves—term-premium compression, MBS spreads, or something else?
  2. How do you account for the interest-income loss to the private sector when bonds are replaced by reserves?
  3. If banks “invest reserves,” by what mechanism—since reserves are stuck at the central bank?
-Jason 

--
You received this message because you are subscribed to the Google Groups "Modern Monetary Theory" group.
To unsubscribe from this group and stop receiving emails from it, send an email to modern-monetary-t...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/modern-monetary-theory/CABuE7%2BMgo5GwC%3DAyx8sVgAZBxYK2O7-k4np5sa1hDHZiDJwVWQ%40mail.gmail.com.

Warren Mosler

unread,
Sep 29, 2025, 9:45:49 AM (6 days ago) Sep 29
to Jay Mills, Bijou Smith, Modern Monetary Theory

Jim Byrne

unread,
Sep 29, 2025, 11:23:42 AM (6 days ago) Sep 29
to Warren Mosler, Jay Mills, Bijou Smith, Modern Monetary Theory
I certainly never said that QE pours money into the economy or give banks funds to lend. In fact, here’s my reply to Bijou.

"In QE the central bank swaps government bonds for reserves. Commercial banks can’t spend those reserves on houses or stocks, but the investors who sold the bonds now hold cash or deposits instead of gilts, and they spend that cash on other assets: pushing up the price of assets like housing and shares, contributing to inequality. I used the word ‘liquidity’ as a short-hand for that story.”

When I said there is more liquidity to spend on things link houses and stocks and shares that’s in the context of assuming MMTers have an understanding of how QE works (which is how I put it). I explicitly say in the podcast - QE is not about dumping money into the private sector. 

What I said has got nothing to do with banks having new funds to lend. 

Bleedin’ ell it’s like the tabloid press deliberately misinterpreting the words of their least favourite politician. 

Jim





Jay Mills

unread,
Sep 29, 2025, 1:20:48 PM (6 days ago) Sep 29
to Jim Byrne, Warren Mosler, Bijou Smith, Modern Monetary Theory



Jim, thanks for laying out your view so clearly. From an MMT perspective, I think we’re actually closer than it might appear. You’re right that QE is a swap of gilts for reserves and that commercial banks can’t “lend out” reserves to households or firms. The effect is to change the composition of private sector portfolios, as you note, not to inject new net financial assets into the system.


Where I’d frame it a bit differently is on the spending channel. When the central bank buys bonds, the seller winds up holding deposits instead of gilts. That substitution can influence asset preferences and prices, but it doesn’t provide new income or net wealth to the private sector — only fiscal policy can do that. To the extent QE shifts portfolios, it can reprice risk assets and reinforce inequality, but it doesn’t add to aggregate spending power in the way a fiscal deficit does.


So I see your shorthand of “liquidity” as a reasonable way to capture that asset-market effect, while also wanting to emphasize that from the MMT side, QE’s limitations are clear: it doesn’t fund bank lending, and it doesn’t expand net financial wealth for households. I appreciate your precision in pushing back on the “banks get new funds to lend” misunderstanding.


Best, 

Jason

On Sep 29, 2025, at 11:23 AM, Jim Byrne <j...@mmt101.org> wrote:

I certainly never said that QE pours money into the economy or give banks funds to lend. In fact, here’s my reply to Bijou.

Jay Mills

unread,
Sep 29, 2025, 1:20:52 PM (6 days ago) Sep 29
to Jim Byrne, Warren Mosler, Bijou Smith, Modern Monetary Theory


Warren Mosler

unread,
Sep 29, 2025, 3:01:03 PM (6 days ago) Sep 29
to Jay Mills, Jim Byrne, Bijou Smith, Modern Monetary Theory
When a gov (through agents) spends, those funds (tax credits) as a point of logic are either used to pay taxes or remain outstanding.
Note that also as a point of logic It can only spend more than it taxes if some agent wants to not spend his income and instead hold it overnight after the close of the business day. 
That is, without a 'prior' savings desire (in real terms) the deficit spending could not have taken place.
The evidence would be the gov offering to buy at ever higher prices with no one willing to sell goods/services in exchange. 

Along with deficit spending, the gov has the option of setting the duration of its outstanding liabilities.
It can offer no securities, in which case the duration will be 0 with the funds that were deficit spent transferable on demand. Or, for example, it could offer, say 3 month bills at auction which, assuming that market clears, the duration of outstanding liabilities would be 3 months. Same goes for the longer term bonds, etc.

And so the question is, since market participants are aware of the duration/yields/etc. before they offer good/services for sale, to what extend do these choices influence selling prices. 

We now have decades of data that indicate that the analysts haven't yet detected any material effect. 

Yet CB's continue to tout it as proactive monetary policy to achieve objectives.

Warren  

Jay Mills

unread,
Sep 30, 2025, 12:58:36 PM (5 days ago) Sep 30
to Bijou Smith, Modern Monetary Theory, Jim Byrne, Warren Mosler
Thanks Bijou for clarifying!

Asset prices seem to go up or down because of: expectations about future earnings and the level of risk investors are willing to tolerate. Stocks rise if companies are expected to make more profits. Bonds rise if investors expect reliable coupon payments and low default risk. If investors demand a lower risk premium because they feel safer, they will also pay more.

Quantitative easing, however, is just an exchange of government bonds for reserves. It doesn’t increase corporate earnings, change future rents, or reduce real economic risks. Since it doesn’t alter those fundamentals, QE itself is not the real driver of asset prices.

Fiscal policy is a genuine driver of asset prices it would seem, because it directly alters private-sector income and risk. When the government spends or cuts taxes, it injects new financial assets into the economy, raising sales and profits for firms and improving expected earnings, which supports higher stock prices. Stronger fiscal flows also reduce credit risk for households and businesses, lowering risk and making investors more willing to bid up assets. 

In addition, government deficits often mean more interest income is paid into the private sector, further supporting saving and wealth accumulation. Historical patterns seem to bear this out: markets surged following the fiscal stimulus of the 2020 CARES Act, while Japan’s long QE programs produced limited asset booms without significant fiscal expansion. 

Asset prices move with earnings, fiscal deficits, risk premium, and government interest payments. QE, in contrast, is just a swap of government liabilities and does not create new net wealth.


Jason

On Sep 30, 2025, at 11:50 AM, Bijou Smith <achrono...@gmail.com> wrote:

👇🏼I agree with you below👇🏼 Jay.  Thanks Warren for laying it out so succinctly. 

For the record: I was asking if anyone knew of any channel whereby QE would have an asset price hike bias. Like you Jay, I was sceptical there was any significant effect.  Nothing to do with bond holders getting out of securities into cash, that's not "QE" in my understanding, so I apologize to Jim if I was confounding the question.  My understanding is QE as a CB policy they believe will stimulate the economy, with low interest rates, lowering cost of borrowing, via "portfolio rebalancing". But most MMT analysis I've read would say that is not going to have the effect they think.  It's more like anti-stimulus bias —  reducing the government deficit. Whether that can drive up asset prices or not I just don't know, and have not seen the evidence. Only vague arguments. Which don't prove anything. Plus, the CB cannot force the financial institutions to sell their bonds, they have to have some other reason, and that's coming from some other policy settings, not QE (portfolio rebalancing)  per se.

Jim later explained his reasoning (correct me if I'm still misunderstanding), which was that the policy driving traders (or funds) to get out of bonds into equities or other investment assets was doing the trick, but that's not QE right?, that's an effect of that other policy. In my mind then that was a very different question (I was guilty of confounding them I guess).  So I still have no clue how QE could shift asset prices. Hence I'm inclined to agree with Warren's assessment, and not just trust the BoE white papers/briefings, they're known to be deeply flawed, quite often.  We could maybe find correlation in time, but you cannot just use that as an argument showing "QE causes" asset price pressures.  I'll concede that maybe I read Jim too quickly, since if he was claiming QE is "all those other policies too" then maybe he's right that the entire pot of policies can cause asset price inflation, i.e., relative price adjustments.  But I just don't see how.  

Jim's other reply to me was about why bond holders would get out of securities, he wrote: "Because the BoE sets a price/conditions in a way that ensures it’s an attractive proposition. The effects of QE is undoubtedly to push up asset price - even the BoE admits that. I’ve already written an entire article on this subject. You’ll find the appropriate quotes there."  
Which I could agree with the first sentence of, no argument. But that's not QE by my understanding, so I'd say the second sentence there is not correct (I am sure Jim's BoE quotes are fine, I just don't trust the BoE, from analogous experience reading my country's CB briefing stuff, the RBNZ, a lot of it also bad stuff, and just sometimes plain wrong).   With the first sentence there (Jim and I) had at that point got onto the portfolio preferences topic, which I would say is a different question to "Does QE cause asset price pressure?", and which I'd agree with Jay & Warren about.

I still think QE does not cause significant asset price pressure, and the BoE white papers saying QE does cause asset price pressure are most likely flawed. I did not know so for sure, hence I was just asking.  But as Jay laid it out, bond trader preferences and the CB policies that motivate them to get out of bonds into risk/assets could I suppose create some relative price adjustments. For socioeconomic inequality if that's relative to wages then it increases inequality. But not via QE, it is via the other policies the CB or parliament sets or changes to motivate those shifts in portfolio preferences which might cause asset price bubbles. That's an investor winners vs losers story, but how is it adding to macro socioeconomic inequality?  

Maybe I'm being dumb? But here ("Portfolio Balance Channel") is an analysis showing "... the shift was mainly into newly issued government bonds." — so, what the heck dudes... it's still the government free lunch causing the increase in inequality.   Basic income for people who already have money in proportion to how much they already have.  That's the driver as I see it, not QE per se.  The research added: "... portfolio rebalancing by fund managers into riskier assets is much smaller in magnitude." (intra-funds investment rebalancing), and most of the rest was "across funds" where there was slightly more going into corporate bonds than new government issue. All told, it seems clear to me it is still the risk free interest-income from the government bonds as the major component of increased socioeconomic inequality.

You can check the BoC reasoning here too: https://www.bankofcanada.ca/2025/02/understanding-quantitative-easing/ — I think it's backwards, would anyone disagree? They are saying signalling interest rates will stay low is their intended effect of QE, and that it'll be stimulus. So, I'd say, it's wrong. But far be it from me to question a central banker to their face!  However, I'll take Warren's view on market price dynamics any day.

Thanks all, and Jim too, for the comments.
:)

Jay Mills

unread,
Oct 4, 2025, 10:36:25 AM (yesterday) Oct 4
to Sanford Friedman, Bijou Smith, Modern Monetary Theory, Jim Byrne, Warren Mosler

Hi Sandy,


Great question — definitely not a dumb one 🙂. From the MMT view, QE is really just an asset swap: the CB takes bonds out of the system and leaves reserves instead. That doesn’t add new net financial assets or boost lending directly, since banks don’t lend out reserves.


You’re right that investors sometimes shuffle into other assets, which can move prices a bit, but real investment in plant/equipment comes from demand and profits, not the reserve mix. So the collateral effect is probably small — but I love that you raised it, because it’s exactly the kind of angle that sparks useful discussion. I’d add that probably deficit spending is the bigger driver in these areas than bond/asset composition  


Best, 

Jason

On Oct 4, 2025, at 10:26 AM, Sanford Friedman <sanma...@gmail.com> wrote:

Hi All

I have been thinking about this from a slightly different angle and I would be interested in this group's input. I fully agree with what has been previously stated, but here is my question.. (assuming that there is not such thing as a dumb question 🙂 as I am no expert)

If QE removes lower quality "bad" assets off private sector balance sheets and puts them on CB's balance sheet - One can assume that the private sector actors likely replace those bad assets with higher quality ones (let's say risk free government bonds) using the new reserves held on their behalf.  To the extent that those higher quality assets could then be used as collateral for private bank lending which in turn might contribute to real asset price inflation (meaning plant and equipment or other hard investments by companies).  I doubt this scenario would have a significant effect but I am wondering if it should be considered in this discussion.

Thanks in advance for your replies.
Sandy
(currently enrolled in Torrens MML program)

Sanford Friedman

unread,
Oct 4, 2025, 11:40:19 AM (yesterday) Oct 4
to Jay Mills, Bijou Smith, Modern Monetary Theory, Jim Byrne, Warren Mosler
Hi All

I have been thinking about this from a slightly different angle and I would be interested in this group's input. I fully agree with what has been previously stated, but here is my question.. (assuming that there is not such thing as a dumb question 🙂 as I am no expert)

If QE removes lower quality "bad" assets off private sector balance sheets and puts them on CB's balance sheet - One can assume that the private sector actors likely replace those bad assets with higher quality ones (let's say risk free government bonds) using the new reserves held on their behalf.  To the extent that those higher quality assets could then be used as collateral for private bank lending which in turn might contribute to real asset price inflation (meaning plant and equipment or other hard investments by companies).  I doubt this scenario would have a significant effect but I am wondering if it should be considered in this discussion.

Thanks in advance for your replies.
Sandy
(currently enrolled in Torrens MML program)

--
You received this message because you are subscribed to the Google Groups "Modern Monetary Theory" group.
To unsubscribe from this group and stop receiving emails from it, send an email to modern-monetary-t...@googlegroups.com.

Bijou Smith

unread,
Oct 4, 2025, 12:23:48 PM (yesterday) Oct 4
to Jay Mills, Sanford Friedman, Modern Monetary Theory, Jim Byrne, Warren Mosler
Agree with Jay, but not a dopey question from Sanford.  The key word is "might" in "... might contribute to real asset price inflation" — so nice you wrote it in italics, kind of answered your own question. How would that cause people to pay more for the "higher quality" assets?  How? The answer is not the QE itself imho. In the initial asset swap the market already presumably had the price and risk factored in to the price they'd be willing to pay. As you sworote, and as I wrote before, some research suggests most of it just goes into roll-over (or "rebalancing"), they purchase new government securities, or corporate bonds, depending on the institution (but someone should check this, don't quote me). This was Bernanke's reasoning anyway, thinking this rebalancing would be a stimulus (kid in the back seat steering the economic car.)

It's also good to always consider both sides of the ledger. When interest rates change, or asset prices move, there's a winner and a loser. But it's the rate of change that matters, not the final positions, since a fiat currency is a gauge system. I suspect a lot of headline rhetoric fails to consider this?  The final position is just the record of account, that, "yep, we created more inequality. Well done boys."  But I don't think the QE is what causes these rates of change. The rate of change is the killer. But as Jay said, it's usually fiscal dominance everywhere, all the time.

Steve Keen recently put out a good analysis for the case of real estate, the other day.  I'm not an expert in markets, but Keen's video is not too long and runs through some of the fundamentals. In short, the Aussie parliament needs no QE at all, they've always been inflating house prices, they use simple "first home buyer" incentives and such-like (fiscal dominance really, even though it's marketed politically as a loan). Both their duopoly political parties. When the government does it this way, it's almost "not even ponzi."  They turn on the tap at the other end of town with government bonds, to feed the rich and the investors, who now can afford the higher house prices (Blackrock et al.).  That's what's causing the massive increases in inequality.... again, just imho.

Thanks for poking this thread, because I don't want to get it wrong. It stifles freedom in writing when I'm uncertain.
-Bij

Jay Mills

unread,
Oct 4, 2025, 12:32:50 PM (yesterday) Oct 4
to Bijou Smith, Sanford Friedman, Modern Monetary Theory, Jim Byrne, Warren Mosler
Good stuff !

Curious - QE just adds reserves to the balance sheet of banks? Banks lose bonds and add reserves. This makes it easier to settle payments and makes it less stressful to extend loans assuming the borrower is in good standing. IOR make those reserves less of an issue since they’re getting a coupon. Other than that can’t see any major effects of QE

Best, 

Jason

On Oct 4, 2025, at 12:23 PM, Bijou Smith <achrono...@gmail.com> wrote:


Reply all
Reply to author
Forward
0 new messages