Treasury market bleg

Here’s Bloomberg:

The Treasury market is now so large that the U.S. central bank may have to continue to be involved to keep it functioning properly, according to Federal Reserve Vice Chair for Supervision Randal Quarles . . .

“It may be that there is a simple macro fact that the Treasury market being so much larger than it was even a few years ago, much larger than it was a decade ago and now really much larger than it was even a few years ago, that the sheer volume there may have outpaced the ability of the private market infrastructure to support stress of any sort there,” Quarles said.

In plain English, what does this mean? What is the bad thing that would happen in the Treasury market if the Fed did not intervene in times of stress? If that bad thing happened, would it create an easy profit opportunity for someone like me (an investor who can wait out periods of stress?)

Why does increased size make the Treasury market more fragile? One normally thinks in terms of size and liquidity being positively correlated. Is “stress” different from illiquidity?

Suppose the Fed intervened in the MBS market during times of stress in the T-bond market. Would that fail to address the problem? I.e., is the problem Treasury debt-specific, and not just a generalized lack of liquidity, (which can be addressed by adding more reserves?)

How does this problem relate to the “safe asset shortage” that people often discuss? Would addressing the safe asset shortage by issuing more T-bonds also make the Treasury market bigger and thus even more fragile?

PS. Just to be clear, Quarles does not seem to be discussing central bank support of government debt in the traditional sense, which meant intervening to keep government financing costs low. When the Fed injects more reserves into the system, they tend to pay interest on those reserves at a rate comparable to T-bill yields. Thus the Fed is not “monetizing the debt” in the original meaning of the phrase.



28 Responses to “Treasury market bleg”

  1. Gravatar of Matthew Waters Matthew Waters
    14. October 2020 at 16:00

    The basic story of past few years is:

    1. Significantly more Treasury issuance in raw dollar terms.
    2. More of the Treasuries being funded through the following chain:

    – Government Money Market Mutual Fund.
    – Repo to broker-dealer.
    – Repo from broker-dealer to hedge fund, or from BD to another BD to HF. HF may make some arbitrage bet. Due to Volcker Rule and other reasons, it can be cheaper to be intermediary vs BD doing bet directly.

    IMO, last few years moved from Treasury buyers being bona fide Treasury buyers who did not transform maturity. Treasuries went to backing money of different kinds, mostly government MMMFs.

    The most basic issue is government MMMFs can face withdraws for taxes. Last Sep had no Fed support of Treasury market at all and top of rates went up to 10%. The “living will” regulations hamstrung banks from tiding over the market while MMMF withdrawals were made to pay quarterly taxes. Living will regs were changed.

    The regulations were changed, but in March there was a huge run *to* bank deposits and government MMMFs. Capital limits were hit and Treasury bid/ask spreads blew out. If no action was done to deal with capital limits, no new financing could have been done for any company or municipality. Even Treasury yields went up a good bit. QE gave capital space to banks through a very dumb method. Capital regs were also changed on 4/1.

    There is zero repo done now, because we’re flush in reserves. Fed has not stopped QE. QE may give more capital space still because banks have to hedge rates and have some capital against rate swaps defaulting. But I think it’s minor.

    In short, Sep-Mar Treasury market support was to control rates. Mar-today support is to require less bank capital against Tsys/reserves and thus allow higher M.

  2. Gravatar of ssumner ssumner
    14. October 2020 at 16:23

    Thanks Matthew. I see how capital limits could be a problem. Other than that, would sufficient reserve injection (in a period of stress) be enough to address these temporary problems? If not, why not?

  3. Gravatar of Benjamin Cole Benjamin Cole
    14. October 2020 at 16:36

    I think Randal Quarles was referring to the possibility that the Federal Reserve would have to be a buyer or seller of last resort for US Treasuries, under certain circumstances.

    Perhaps. My own view is there is a near bottomless global market for US Treasuries, and some other sovereigns. There may be fleeting episodes that require massaging.

    There also appears to be a near bottomless market for US cash (digital or otherwise). As of now, this means the Federal Reserve can transfer US liabilities (debt) back onto the US asset sheet (the Fed’s balance sheet) at any time, through quantitative easing.

    It is heresy to say so, but the US government can print money to run government operations presently, without much inflationary effect. It is wrong, wrong, wrong, and wrong but it is also the truth.

  4. Gravatar of Matthew Waters Matthew Waters
    14. October 2020 at 17:09

    Yes, new reserves kept market functioning in both September and March.

    QE had more potency in March than usual in a very strange way. The big banks were at Supplementary Leverage Ratio, which includes both Treasuries and reserves in the denominator. So QE should not free up capital, since reserves are also in SLR. However, the new reserves were used to refund repos from government MMFs and MMFs used Fed’s reverse repo. Reserves were no longer on balance sheet of a bank, freeing up bank to buy Treasuries or make loans.

    Reverse repo spiked to $280bil. After Treasuries and reserves were excluded from SLR, reverse repo dropped backed down.

    To be honest, it’s tough to understand Quarles comments as they apply to post-4/1 markets. The spreads tightened again according to Figure 3 here.

    The SLR exclusion ends next year supposedly. With capital limits, QE will again alleviate stress through reverse repo. Capital reqs could also have some countercyclical leeway to not create hard limits on M. Assuming equity is impossible to raise in stressed times, total deposits cannot go up with simple capital ratios.

  5. Gravatar of Michael Sandifer Michael Sandifer
    14. October 2020 at 19:32

    Would a permanent repo facility and/or allowing any expansion in number and type of counterparties help?

  6. Gravatar of Sean Sean
    14. October 2020 at 19:46

    It’s not a problem of the market being too big. If the profits were there then more capital would come in.

    The issue is a lot of the medium term liquidity moved into risks sensitive actors. It’s ran by the citadels millenniums etc that may be well capitalized but run risks very tight. The deal for those funds with their investors (pension funds) is their not allowed to lose money which gets them investors willing to accept 8% returns in exchange for limiting risks.

    When stress like covid hit those guys turn into sellers of positions. So the marginal buyer turns into a marginal seller and there’s not much to step in short term. Bank prop desks could just ignore Mark to market for a while.

  7. Gravatar of Nick S Nick S
    14. October 2020 at 22:27

    Summer – I think you’re missing the effect of simple supply/demand dynamics….

    You asked… “Why does increased size make the Treasury market more fragile?”

    As the govt goes deeper and deeper into debt, they need to finance it via increased TSY supply. As a result, rates will sell off without Fed intervention. This intervention could be in a few different forms…

    1.) Fed starts doing reverse repos to make purchasing the TSY’s an attractive positive carry position for investors (The definition of the reverse repo program is misunderstood by most and even defined incorrectly on the frb website. This activity involves the Fed making loans to investors which are backed by the TSY bonds that the investor is purchasing and posting to the Fed as collateral. This is an asset on the Feds balance sheet and simply a loan that has recourse to the investor’s TSY bond collateral in the case of default). This fed program injects reserves into the banking system, driven by the sale of TSYs in the primary market initially to broker/dealer member banks which cross the securities to investors who obtain financing for the purchase via the fed reverse repo facility.

    2.) Fed purchases TSYs outright from broker/dealer member banks, injecting additional excess reserves into the banking system

    You also said that the Fed is not monetizing the debt because “they tend to pay interest on those reserves at a rate comparable to T-bill yields.” This is false… as the fed finances the interest payments on the excess reserves via freshly printed money. Before you counter this point by claiming that the Fed is actually financing the IOER payments using the interest earned on their T-Bill assets, which is paid by the US Treasury… I contend that the US Treasury funds these payment obligations via the issuance of additional debt, which is again purchased by the Fed via money printing… as we know that US govt tax revenue is not even close to being large enough to service debt payments.

    So yes… increased size of the TSY mkt absolutely makes it more fragile. Even though there will always be a last resort buyer (the Fed), the Fed needs to resort to purchasing these TSYs via money printing, increasing the money supply… and at the end of the day, TSYs are IOU dollars…so why would anyone want to buy a financial instrument that returns you dollars at a future point in time when there will be a much larger supply of dollars that are worth less? There is no free lunch here. The Fed can keep the price of TSYs high, but at the expense of a weakening dollar (check out the weakness in the DXY lately)

    Mark – unless I’m misreading your comment, I believe you’re mistaken on your characterization of money market fund activity… as these MMF’s are typically the largest buyers of reverse repo assets. When they receive redemptions, the supply of TSY financing dries up (less demand for reverse repo assets) which causes rates to sell off…. hence the need for Fed intervention and establishment of their repo facility.

  8. Gravatar of Ray Lopez Ray Lopez
    15. October 2020 at 04:41

    Good column. Note Sumner does not pretend to know the answers. It’s all up in the air. Ben Cole’s idea of MMT could be bogus, as even Ben seems to agree. The dollar could crash and no longer be the de facto world currency (though what will replace it?) Analogous with stocks, it’s the escalator up, and the elevator down. I recall when my 1% family bought Citigroup (C) in fall of 2007, it was touted by Morningstar as a blue chip, then within a year or so it lost 90% of its value. Same could happen with the US dollar. Got gold?

  9. Gravatar of Michael Rulle Michael Rulle
    15. October 2020 at 06:54

    It seems obvious, on the one hand, that the Fed does not believe the public would own all treasuries at these rates. Otherwise, why do they own them? What if they owned 10% of what they do now? Excess reserves would decline—-but would liquidity decline? This seems to be what Quarles is implying—I think the Fed agrees so they are providing liquidity—-and we seem not to have a current monetary crisis.

    It is like being the lender of last resort—to the Treasury. One thing that seems different is the magnitude of the increase in Treasury debt. According to FRED total Federal Debt as a percent of GDP increased by 60% from 9/2008 to Q2 2019. From Q2 2019 to Q2 2020 another 35%. At Q2 2008 Debt to GDP was 65%. Now, it is 135%.

    Arithmetic–no interpretation. But I think if we had played out such a scenario circa 2000, we would have thought this would have been a disaster. But it is not–or so it seems.

    Is this an illusion because of the excess reserves offset?

  10. Gravatar of ssumner ssumner
    15. October 2020 at 08:17

    Everyone, There’s a lot of talk about the risk of interest rates going up. First of all, high interest rates are not the current problem facing America. And rates typical fall sharply in times of market stress, so I don’t see evidence for your claims. Second, why would higher interest rates be a problem? Isn’t that how markets are supposed to work? (I’ll ignore your answer if it involves reasoning from a price change)

    I do understand that there may be times when the Fed needs to inject reserves because reserve demand spikes—that’s just ordinary monetary policy—EC101. I still don’t see what’s so special about this situation.

    Matthew’s comment on top seems best. Perhaps that are capital requirement problems here. But I’m skeptical that the size of the T-bond market is the underlying problem. What happened to the safe asset shortage?

  11. Gravatar of Brian Brian
    15. October 2020 at 11:08

    But for a while rates went up when the market was stressed. 10 year Treasury March 9 was 54 bp and on March 18 it was 118 bp.

  12. Gravatar of Brian Brian
    15. October 2020 at 11:20

    Low asset prices were a problem…

  13. Gravatar of Brian Brian
    15. October 2020 at 11:28

    This memo came out after yields went the wrong way for a few days…

  14. Gravatar of Brian Brian
    15. October 2020 at 11:43

    “Safe asset shortage” when people have a high level of stress means yields have been decreasing in recent days. “Safe asset shortage” when people have an extreme level of stress means they want to sell everything including bonds and hold cash. Cash is the safe asset. In the second case yields can increase.

  15. Gravatar of Brian Brian
    15. October 2020 at 11:58

    Your comment… “Perhaps that are capital requirement problems here.”

    Sounds like a euphemism for forced liquidation. Forced liquidation is a euphemism for a condition of low asset prices for a leveraged investor which in turn seems like reasoning from a price change which you are asking us not to do. How am I wrong here.

  16. Gravatar of sean sean
    15. October 2020 at 12:07

    I don’t believe hes referring to level of interest rates.

    But trading activity. How much liquidity is in the market. Functioning of spreads between various treasury products – on the run, off the run, futures basis. If the market can’t properly arbitrage these products then its not going to function at pricing credit and equity risks either. Or providing capital to the real economy. Treasury market was the first thing the fed fixed this crisis.

  17. Gravatar of dtoh dtoh
    15. October 2020 at 15:51

    Liquidity can and should be defined and measured as the spread between the bid and offer prices for an asset.

  18. Gravatar of Benjamin Cole Benjamin Cole
    15. October 2020 at 16:27

    Ray Lopez: When it comes to monetary policy, if you are not confused, probably that’s because you do not understand the facts.

    When it comes to monetary policy, the most prominent thinkers often have diametrically opposed views and perspectives.

    Some highly intelligent participants in macroeconomic policy-making, such as Martin Feldstein and Paul Volcker, were wrong on the pending direction of interest rates and inflation for 40 years in a row. I assume they left their final, and unchanged, thoughts on the topic on their headstones.

    I have absolutely no doubt, and I am absolutely convicted of the opinion, that interest rates and inflation should have been much higher over the past 40 years, and the dollar should have collapsed!

    I am an acolyte in the Temple of Conventional Macroeconomic Theology.

    By the way, did you know free trade is divine also?

  19. Gravatar of ssumner ssumner
    15. October 2020 at 19:54

    Brian, Thanks, I forgot about that brief spike. Of course in absolute terms even a 118 basis point yield is quite low. So in what sense was this a big problem?

    You said:

    “Safe asset shortage” when people have a high level of stress means yields have been decreasing in recent days. “Safe asset shortage” when people have an extreme level of stress means they want to sell everything including bonds and hold cash. Cash is the safe asset. In the second case yields can increase.”

    This is interesting. So can these problems always be solved with more reserves? Or are there cases where more reserves don’t help, only intervention in the T-bond market works?

    dtoh, That definition seems reasonable. How big did those spreads get in March?

    Just to be clear, I sometimes follow the press in referring to reserve injections meeting a demand for “liquidity”, in which case I actually mean a demand for base money.

  20. Gravatar of Ray Lopez Ray Lopez
    15. October 2020 at 19:55

    @Ben Cole – I agree with you. Martin Feldstein and Paul Volcker were overrated, the latter taking credit for ‘breaking the back of inflation’ when in fact, like the bank panic of 1933, the public’s mood simply switched and inflation burnt out on its own. There was, during the early 1980s era, even a short period when the Fed actually cut rates and inflation fell (yes, you read that right, of course you can make your priors work if you assume lags, but the cleanest way of squaring the circle is to assume money is short-term neutral). As for free trade, it works on paper but in practice it’s observed in the breach. The US north got by just fine with protectionism, while the US south suffered under free trade in agriculture (as has Africa, Asia). As for advocates of free trade, they are mouthpieces for industrialists, by and large. BTW I’m for free trade–in services. Let the Third World come to the USA and do everything under the sun, including dentistry, as done in the PH, and with unlicensed street dentists, lawyers, ‘doctors’, caveat emptor.

    The rule of thumb in all things human is human nature: “I have been a selfish being all my life, in practice, though not in principle.” – Jane Austen, Pride and Prejudice.

    PS–for fun, as he has no chance of winning, I might vote for Trump in Virginia. A protest vote, albeit despite his presidential incompetence, Trump is looking out for my 1% interests more than tax-and-spend Biden. MAGA!

  21. Gravatar of Nick S Nick S
    15. October 2020 at 21:18

    Sumner- Have some comments on your points below…..

    “ First of all, high interest rates are not the current problem facing America.”


    “And rates typical fall sharply in times of market stress, so I don’t see evidence for your claims”

    -Correct, but housing prices “typically” always go up too though right?

    “ Second, why would higher interest rates be a problem?”

    -Correct, higher rates wouldn’t be a problem in the long run, because higher rates would trigger the economic reset (destruction/default of unproductive capital investment…I.e. the proverbial “Zombie” companies) required to reassume a sustained path of real growth. Higher rates would simply be a consequence of the Fed ceasing to manipulate the market for govt debt, which would also force govt to rein in unproductive public programs and agencies that are currently financed with debt.

    “Isn’t that how markets are supposed to work?”

    -Partially correct… FREE markets work. The current “market” for Treasuries (as well as many other asset classes) is severely manipulated by the Fed.

    “But I’m skeptical that the size of the T-bond market is the underlying problem. What happened to the safe asset shortage?”

    -The trendy talk of a “safe asset shortage” in academic circles is riddled with problematic logic and misleading assertions….

    1.) The designation of an asset as “safe” is extremely subjective and 100% contrary to the more developed academic theory of risk neutral pricing.

    2.) Is buying “safe” US Treasury debt that currently returns a negative real return “safe?” Not to mention the potential for MTM losses should rates sell off…

    3.) If there was really a “safe asset shortage” the Fed wouldn’t need to intervene in treasury markets, as investors in the market would provide enough demand to drive rates low enough to meet the Fed’s target. This begs the question, are treasuries actually “safe?” The recent plunge in the US dollar index suggests that perhaps they are not.

    “ Perhaps that are capital requirement problems here.”

    -If capital requirements are forcing banks to maintain much larger balances of “safe assets” (treasuries, agency mtge, dollars) than normal, why are their CDS spreads and equity valuations selling off so much relative to other sectors? If the bank’s asset base is increasingly composed of “safe assets,” this implies that bank debt is increasingly collateralized by “safe assets,” in which case, bank credit spreads should rally if there is truly a “safe asset” shortage…. the opposite has happened.

  22. Gravatar of James Alexander James Alexander
    15. October 2020 at 22:46

    Those leverage ratios (capital/assets) are a bit of a menace in bad times. The massively complex risk-weighted leverage ratios and the simple unweighted ratios, both. They were designed to be used in good times to prevent banks “cheating” by inflating balance sheets to get more (or at least sustain) profits at lower margins and lower margins, while increasing (or at least sustaining) a return on that capital. In bad times as cash returns to banks, especially some ultra safe banks, balance sheets have to inflate. But they can’t inflate unless leverage rules are relaxed.

    And then … as cash returns to banks they have to place the cash somewhere safe. Treasuries are risky due to mark to market issues as comments point out. So, the Fed has to get increased reserves to absorb that cash that has to be parked by the banks somewhere.

    I think!

  23. Gravatar of Benjamin Cole Benjamin Cole
    16. October 2020 at 02:09

    Special to Ray Lopez:

    Most MI readers probably know this, but worth reading.

    The question is: What is China doing right, and can they sustain? Can any part of their success be mimicked?

    The simple chant that “China is liberalizing” is true but with huge limits. The government owns all the land, heavily subsidizes industry, limits consumption, follows five-year national plans, runs the PBoC and the financial sector, controls all media and discourse.

    But the CCP is successful economically, growing through the 2009 debacle, and again through the C19 debacle.

  24. Gravatar of ssumner ssumner
    16. October 2020 at 09:00

    James, That all sounds reasonable. I read that as implying that it’s reserves that really matter, not Treasuries.

  25. Gravatar of Spencer B Hall Spencer B Hall
    16. October 2020 at 10:07

    re: “but in March there was a huge run *to* bank deposits and government MMMFs. Capital limits were hit and Treasury bid/ask spreads blew out.”

    People don’t know a debit from a credit. That’s called disintermediation of the nonbanks (representing a destruction in the supply of loanable funds). Same thing happened last Sept.

    The “shortage of safe assets” simply represents an excess of savings over government targeted real investment outlets. As N-gDp has fallen, so have investment alternatives (just as predicted in the late 1950’s).

  26. Gravatar of Spencer B Hall Spencer B Hall
    16. October 2020 at 10:23

    The economies of the world are being run in reverse, increasing income inequality, decreasing standard of livings.

    Banks are “Black Holes”. They impound savings. Banks don’t loan out deposits, banks create deposits when they lend/invest.

    Savers never transfer their savings out of the banking system (where all monetary savings originate). The only way to reduce bank deposits is to hoard currency or convert to other national currencies.

    The typical way to reduce the volume of bank deposits is for the saver-holder to use his funds for the payment of a bank loan, interest on a bank loan for the payment of a banks service, or for the purchase from their banks of any type of commercial bank security obligation, e.g., bank stocks, debentures, etc.

    Savings transferred through the nonbanks (financial intermediaries), activate savings. But there is just an exchange in the ownership of pre-existing bank deposits (a velocity relationship). There is no change in the volume of bank deposits, nor the bank’s earning assets.

    It’s raw accounting — stock vs. flow. You have to retain cognitive dissonance capacity, like Walter Isaacson described Albert Einstein’s ability:

    to hold two thoughts in your mind simultaneously – “to be puzzled when they conflicted, and to marvel when he could smell an underlying unity”.

    Take the “Marshmallow Test”: (1) banks create new money (macro-economics), and incongruously (2) banks loan out the savings that are placed with them (micro-economics).

    See Steve Keen: “Banks don’t “intermediate loans”, they “originate loans”.

    It is incontrovertibly the single biggest error, and longest standing error, in modern human history.

    Economists are one dimensionally confused. Savings flowing through the banks increases the supply of loanable funds, but not the supply of money (a velocity relationship). Where do you think velocity has gone?

    Link: Dr. Philip George: “The velocity of money is a function of interest rates”

    As Dr. Philip George puts it: “Changes in velocity have nothing to do with the speed at which money moves from hand to hand but are entirely the result of movements between demand deposits and other kinds of deposits.”

  27. Gravatar of James Alexander James Alexander
    16. October 2020 at 11:40

    Yes. Reserves not treasuries.

    And if the Fed is so useless that it refuses to generate NGDP growth and inflation/NGDP growth heads down like it is in the Euro Area then treasuries are far worse than cash. Returns on cash can’t ever go negative, in nominal terms (of course), unlike treasuries. Cash is THE benchmark for all marks to market.

    When treasury yields go negative you are guaranteed a (nominal) loss if you hold to maturity. Not on cash.

  28. Gravatar of Postkey Postkey
    16. October 2020 at 12:56

    ‘An empirical test is conducted, whereby money is borrowed from a cooperating bank, while its internal records are being monitored, to establish whether in the process of making the loan available to the borrower, the bank transfers these funds from other accounts within or outside the bank, or whether they are newly created. This study establishes for the first time empirically that banks individually create money out of nothing. The money supply is created as ‘fairy dust’ produced by the banks individually, “out of thin air”. . . .
    Thus it can now be said with confidence for the first time – possibly in the 5000 years’ history of banking – that it has been empirically demonstrated that each individual bank creates credit and money out of nothing, when it extends what is called a ‘bank loan’. The bank does not loan any existing money, but instead creates new money. The money supply is created as ‘fairy dust’ produced by the banks out of thin air.32 The implications are far-reaching.’

Leave a Reply