Skip to content

Ep. 68 Jeff Snider Explains the Repo Market Flare-Up, the Fed’s Phony Solution, and the Global Dollar Problem

Jeff Snider is Head of Global Research at Alhambra Investments. He talks with Bob about the recent spike in lending rates in the repo markets, and how the Fed’s attempted solution fails to address the real problem. He then relates the repo problem to the global monetary system, which has suffered from major imbalances going back to 2007 that have yet to be corrected.
Mentioned in the Episode and Other Links of Interest:

The audio production for this episode was provided by Podsworth Media.

About the author, Robert

Christian and economist, Chief Economist at infineo, and Senior Fellow with the Mises Institute.


  1. Peter on 10/17/2019 at 10:28 AM

    So if an institution swaps treasuries for cash at the repo window, takes the cash, buys new treasuries, swaps those for cash, use that cash to buy treasuries, etc.
    How is that not money printing and monitising debt?

    • Tel on 10/18/2019 at 11:38 AM

      I agree … it’s a logical impossibility for the Fed balance sheet to expand without new money being created.

      Suppose I start with 100 marbles of various colours and spread them out on the pavement in a number of circles, then I let the kids exchange marbles between circles back and forth … what you discover is that NO new marbles get created in this process. If new marbles appear on the pavement then it certainly was not the side effect of an exchange operation.

      Suppose the Fed buys a million dollars worth of assets, that in itself is an exchange, but now show me where someone paid a million dollars into the Fed beforehand in order for the Fed to have money to buy new assets. They print money in order to buy those assets, because there never was a depositor who put that money in.

      Having said that … there’s a big practical difference between circulating money in the general economy and money simply parked at the Fed in the form of reserves. It’s the same money, but doing different things. With large reserves in the system right now, it isn’t circulating, so the new money has been created but those marbles are in people’s pockets and out on the pavement only a small handful of extra marbles are appearing. That’s OK though, the USD is strong, but not too strong and price inflation is being delivered very much like the Fed said it would do.

      Peter Schiff is predicting a US dollar collapse and he has been predicting that for quite some time, but I have Australian dollars in my hand and if you think that holding cash in USD is ugly, you are kidding yourself … my cash in AUD is down 30% in value over the past few years. You Americans should be happy with your Fed … when you compare with the alternatives out there. The Australian RBA has cut our rates to 0.75% which is the lowest in recorded history for this country. If it was up to me, I would fire everyone at the RBA and hire Janet Yellen tomorrow.

      Finally, I’m predicting that the Fed will open up a bit of a spread:
      * Overnight “headline” rates will stay where they are or got up a little.
      * Interest on mandatory reserves will go down a touch.
      * Interest on “excess” reserves will go down a bit more.

      This spread will liberate some of that money that was created but sits on the sidelines. Not sure what we are talking about, maybe 50 basis points? Enough to do the nudge nudge thang.

      • Peter on 10/19/2019 at 1:44 PM

        Yes, I am from the ECB plantation and it is even worse here. Greece rulers can virtually ‘borrow’ at zero.
        Like you said,no one restrained from consumption because they deposited at the fed. The fed just reduced risk profile by swapping failed investments by cash. The central banks collect failed investments. That is an incentive that will motivate them to make more risky investments. That will happen when the central banks reduce interest on excess reserves.
        They will loan them out again in anticipation of the next swap with the central banks. Profits are theirs, losses are for the fed. Because these lossy assets are ‘illiquid’, still valuable, but they can’t sell them somehow. If you can only sell them to a guy with a printing press, you cant sell them at that price.
        Days go buy no Japanese governments bonds are traded. Its a stealth nationalisation of the economy. I think it will fail as people eventually will rise up when their currency loses value. You already see riots now in Hong Kong, Barcelona, Paris, Chili, Ecuador, farmer protests in the netherlands. The rulers will have to increase oppression untill it bursts.

    • Martin Brock on 11/01/2019 at 12:50 AM

      The institution swapped cash for the Treasuries first, i.e. it purchased the Treasuries. If it then swaps Treasuries for cash the in the repo market, it has the cash again, but it also has an obligation to repurchase the Treasuries when the repo contract expires, so it only has the cash temporarily. If it uses the cash to purchase more Treasuries and swaps the additional Treasuries for cash again, it now has an obligation to repurchase twice as many Treasuries, but it doesn’t have twice as much cash. Cycling the cash through Treasury purchases as you describe is like borrowing money to buy real estate and then using the real estate as collateral for another loan.

      • blunthumb on 04/26/2020 at 11:57 PM

        I believe Peter was illustrating a situation where, for example:
        JP Morgan engages in a repo agreement with Black Rock. BR gives JPM $1M UST for $1M USD.
        JPM takes the $1M UST to the repo window and gets $1M USD.
        JPM engages in a repo agreement with Berkshire Hathaway. BH gives JPM $1M UST for $1M USD.

        On the other side,
        BR comes back at the end of the term and pays JPM $1M +.8% APR USD. JPM uses $1M UST from BH to pay BR.
        JPM takes $1M + .25% APR to repo window to repurchase $1M UST.
        BH repurchases the $1M UST for $1M + .8% APR USD.
        JPM ends up with original $1M + 1.35% APR in interest accrued over the term.

        I believe it would be more accurate to say that the debt is being monetized for the duration of the term of the repo agreement.
        If the counterparty defaults on their repurchase obligation with the fed, I think that would definitionally monetize the debt for a duration beyond the terms of the repo agreement.

  2. […] The graph below is the difference between the Fed Funds Rate and IOER during the last few months. It is clear that the force pushing the Fed Funds Rate above the IOER is getting stronger.3 […]

  3. Aryslan on 10/17/2019 at 4:17 PM

    Excited for this episode. Listened to this guy on MacroVoices podcast.

  4. Joseph M Stivaletta on 10/17/2019 at 5:20 PM

    So may it be reasonable to assume that we be in deep shaite because the world economy does not use a sound monetary system? I find it amusing that people waste so much time trying to understand the problem domain of a system in an attempt to fix a basically unfixable system.

  5. Lysander Spooner on 10/18/2019 at 5:55 AM

    Is trying to fully understand the repo crisis a fruitless exercise?

    Surely the lesson is that the Austrians are right:
    – all knowledge is local & therefore no one person can ever know everything
    – the calculation problem means that it is impossible to ‘manage’ any market & the money market is no different

    It doesn’t really matter what will trigger the final collapse of the system. Indeed it is safe to predict that after the system collapses economists and historians will argue for eternity what precise mechanism caused the problem.

    What matters is that the current attempt to manipulate money seems to be shuddering to a end. As Bob says in the intro song “there’s a tidal wave coming, what you gonna do?”. I just wish I had a better plan than keep going until the crash & then hope for the best…

  6. Potpourri on 10/19/2019 at 12:41 AM

    […] In the latest Bob Murphy Show I interview Jeff Snider on the repo market […]

  7. Greg on 10/19/2019 at 4:03 AM

    Great timely discussion. Thank!

  8. David Colucci on 10/19/2019 at 5:26 PM

    This episode is GOLD

  9. David Colucci on 10/19/2019 at 5:32 PM

    This episode is gold…., conversation from ZH,…

    You have it ,combined with a buyers strike on USTs that leaving the PDs way overweighted with bonds that no one wants to touch at these low interest rates.REPO was/is saying it needs to be at least 10% on the 10 year for any private takers for the now junk bond USTs.

    Its about to get much worse for the buyer of last resort IMO.



    33 minutes ago
    Bang. Boom. A runaway interest rate on interbank swaps would literally stop the debt economy in its tracks. This angle is so interesting to me because it shows the interconnectivity of actual assets to money. They can print Money, but they can’t produce assets, and when the system is debt based it is collateral based. The collateral reality, in a sense, destroys the theory that increasing money supply will endlessly increase prices ;at least when the currency enters the system as debt.



    24 minutes ago
    So. Is it inflationary or deflationary. Does this put upward pressure of the interest paid to holdings at the FED? Does this start to have that pent up liquidity leak into the system?

    • Tel on 10/19/2019 at 9:37 PM

      That reminds me, there’s no such thing as “safe assets”. In particular, government bonds cannot be considered safe assets.

      In the most obvious sense … government bonds are printed certificates, no different to paper money when you get to the heart of it. As you point out, anything that can be printed cannot also be a real asset. Using a government bond as “collateral” for a loan is like painting a picture of stairs on a wall, then trying to walk up the wall.

      Physical assets cannot be 100% safe, because physical things are degradable. A house might burn down, a vehicle might crash, a business could fail and go bankrupt. Even a gold bar might get stolen or lost, and anyway the price of gold depends on what someone will pay for it, so there is risk in all asset price volatility. That’s OK because humans are built to assess risk and use some judgement … we do it every day when walking on a slippery wet floor (statistically the most dangerous thing most First World citizens encounter).

      But debt based assets are also not 100% safe. All financial assets are based on the concept of an IOU note … which must necessarily have a counter party. Suppose I lend you money on a mortgage with your house as collateral. Yes that’s a physical asset, but I actually don’t want the house, I would prefer you to work for me paying interest. The house is merely some additional leverage … in case you get reluctant to make those payments. If I’m holding mortgage debt it might seem like I’m betting on the real estate prices, but no I’m really betting on you who lives in that real estate. If you lose your job and can’t pay then I’m in trouble.

      Let’s move along to government bonds as a financial asset … suppose I purchased a 10 year government bond, who is the counter party? Is it government? No … government cannot be a counter party because they don’t produce anything valuable. Governments offer violence as their market product … but I don’t have a use for that … what I’m hoping for is that government directs its violence at other people, in order to make them work for me and pay back the bond.

      Putting that another way … future tax payers are the counter party … because it’s future tax payers who produce something that I might find valuable. Government is an intermediary between the bond holder and the future tax payers. I’m betting on those tax payers being in good shape to pay me back. I don’t even know who those people are (it might be ME who gets taxed to pay back my own bond) but I can roughly guess that a future healthy economy with high employment, good wages, high productivity is probably also likely to deliver some decent tax base.

      However, a struggling economy, with poor employment, and limited surplus, cannot also sustain high taxes. In specific terms, the worst situation for bond holders would be a “Stagflation” scenario where prices are going up and employment is going down. So right now the “10-Year Treasury Constant Maturity Rate” is very low by historic standards. It’s dipped like this twice before:
      * July 2012
      * Sept 2017
      Now it’s dipped again … so if you buy a 10-Y bond now, the interest rates can only go up from here … and you lose money probably within a year. As interest rates go up, the resale price of your bond goes down. It’s a terrible investment because the bond rate is not even sufficient to keep up with price inflation so holding it you lose and selling it you lose. If “Stagflation” hits and interest rates keep rising without real-world productivity then the bondholders are in for some terrible losses. That might seem unlikely, but it’s not impossible … ignore anyone talking about “safe assets”.

  10. will on 10/23/2019 at 1:35 AM

    Where’s the money printing..? When the fed buys something with money it doesn’t have — that’s money printing. Hello? like duh?

    Or did the fed only buy things with money it had obtained legitimately? lolol

  11. […] ep. 68, interview with Jeff Snider on Fed’s repo […]

  12. Martin Brock on 11/03/2019 at 1:21 PM

    I listened to this podcast several times, because the subject seems very important, but it raised more questions in my mind than it answered.

    A repurchase agreement requires the seller of a bond to repurchase the bond after a period of time, typically a matter of days, at a price specified in the contract, so the buyer of the bond essentially lends money to the seller with the bond as security, and the difference between the repurchase price and the purchase price is effectively an interest payment. Right?

    Does the contract also require the buyer of the bond (the “lender”) to sell it back at this price, or is the repurchase optional for the buyer? If the buyer need not sell the bond back at the contractual, repurchase price, the buyer potentially gains if the price of the bond rises during the contract period; otherwise, the seller retains this upside potential.

    If a bond (like a Treasury) has no default risk, the risk of this rising price is the only risk factor? This risk could only justify a lower interest rate, never a higher rate?

    Snider asserts a shortage of collateral for repos. Since Treasuries are nominally riskless, a shortage of Treasuries (relative to dollars chasing Treasury repurchase agreements) can only lower the interest rate? This point is most confusing. A shortage of collateral seems to imply a lower rate, not a higher rate.

    If we’re discussing collateral other than nominally riskless bonds, the story changes completely. The risk of default then dominates the repo rate, and the supply of collateral hardly matters. Snider does assert a shortage of “quality collateral” and alludes to swaps of dicey bonds for Treasuries preceding repo agreements. These dicey swaps might subject the Treasury involved in a repo to a prior claim, i.e. the collateral might be fraudulent? The collateral in a repo could be fraudulent for any number of reasons. Treasuries could be stolen or counterfeit. Is fraud, or a particular pattern of fraud, entirely responsible for the spike in repo rates?

    Apparently, Snider believes that the repo market is fragile, because institutions needing liquidity are sitting on far more dicey assets than “quality” assets. Even if the Federal government increases the supply of Treasuries, the institutions able to purchase these Treasures need not be the institutions needing liquidity in the repo market. As Snider says, if the Treasuries only end up on the Fed’s balance sheet, corresponding to increased reserves for banks not seeking liquidity in the repo market, increasing the supply of Treasuries has no effect on the market.

    So the problem is that zombie banks are still zombie banks and will remain zombie banks until a bankruptcy court finally shoots them in the head? By intervening in a shadowy repo market, the Fed is only conspiring to hide this risk? Shareholders in these banks are at risk? Depositors are at risk? Which banks? That’s the sixty-four trillion dollar question.

    • Dennis Foster on 11/05/2019 at 5:00 AM

      I agree that there is more than a little confusion here. I don’t understand why holders of excess bank reserves don’t jump at the chance to lend these out in the repo market at high interest rates (5%-10%) for the short term. I understand the idea that there is a shortage of buyers, but I don’t understand why. And what does a “shortage of collateral” have to do with a shortage of buyers? That would imply a shortage of sellers, and have an opposite effect on interest rates as you note.

      I do understand the argument that a deterioration in the quality of the collateral would tend to raise rates, but we’re talking about UST’s here and not MBS’s or some other instrument as best I can tell. The notion that this market was most impacted in 2008 (by the withdrawal of MBS as acceptable collateral) begs the question of why these things are happening now? And why if the Fed has been doing some unwinding of its balance sheet? It seems quite counter-intuitive.

      All I seem to get from this is that banks are just not willing to lend out their excess reserves. Calling them part of an “asset swap” seems misleading. [Yes, banks have traded bonds for reserves, but that is what liquidity means and that provides the buying side of the repo market in the first place.] Snider asserted that this was also happening with the Fed’s recent repo actions – their buying spree ended up sitting on the sidelines in the accounts of the primary dealers and not injected into the market. So, if the Fed had done nothing then the two day spike would have ended?

  13. malika on 11/14/2019 at 7:08 AM

    Great point ,good thanks .

Leave a Comment