In shadow banking amongst primary dealers & large institutions...I understand repurchase agreements function similar to deposit accounts (Gorton, et al). Additionally, I understand how traditional commercial banking produces M1 in the form of deposits, etc.

My question is: Where is the analogous money being created by shadow banks? I.e. I cannot use a Repo agreement in a transaction, unlike how I can use deposit accounts in commercial banks. Intuitively I feel money is being created, but not in the exact transaction-useful form as traditional banking.

Is the "money" being created effectively by virtue that the collateral can be rehypothecaed?

To paraphrase Jerry Maguire...Show me the transaction-useful money!

  • $\begingroup$ I don't understand the question. Are you talking about Fed repo operations as part of QE? In this case it's the fed producing the money and buying the security, but just allowing the bond to mature or otherwise ignoring the sell back (or reverse) $\endgroup$ – Frank Apr 2 '20 at 21:09
  • $\begingroup$ Hey, ya maybe I didn't give enough context. So, Commercial Banks produce short-term debt (deposit accounts) which add net NEW money to M1, correct? Repo functions VERY similarly, but for the non-commercial banks (i.e. 'Shadow Banks') -- So my question is: Is there net "new" money being created by Repo? And is re-hypothecation of collateral somehow related to this "new" money? (Re-hypo == Using your posted collateral for my own purposes) $\endgroup$ – Davis Clute Apr 2 '20 at 22:08
  • 1
    $\begingroup$ Can you add a link to "Gorton, et al", or at least name the paper or define "transaction-useful money"? I get zero google hits for this term. $\endgroup$ – Fizz Apr 3 '20 at 5:09
  • $\begingroup$ Also note that there's no standard definition for "M1" or all the M-terms for that matter. See en.wikipedia.org/wiki/Broad_money $\endgroup$ – Fizz Apr 3 '20 at 5:37
  • 1
    $\begingroup$ "Transaction-useful" I just used for shorthand. My point was: if un-conventional/shadow banking produces money (like its commercial counterpart), is it useful in transactions? I'm not sure MMF claims or Repo agreements are. But does that mean it's "less" money-like if can't be used directly in a transaction? $\endgroup$ – Davis Clute Apr 3 '20 at 17:51

Well, the ECB has given it [re]consideration... by excluding repos from their M3 starting in August 2012.

From a monetary analysis perspective, it should be checked, then, whether any of the liabilities issued by the shadow banking system substitute traditional bank deposits – that is, whether they are comparable, say, with the bank deposits included in M3 in terms of their degree of liquidity and their risk characteristics.[footone: The extent to which these liabilities are used as money substitutes is subject to cyclical fluctuations and dependent on market developments.] If this is the case, increasing shadow banking activity would diminish the meaningfulness of monetary indicators, with the monetary aggregates needed for price determination in the goods market, in particular, being defined too narrowly. That is why, in an ever-evolving financial system, it is crucial to regularly review the definitions of monetary aggregates. At the launch of EMU, the Eurosystem decided that the group of monetary financial institutions (MFIs) designed to capture the money-issuing sector should include not just euro-area commercial banks but money market funds as well, the rationale for this move being that for investors, money market fund shares were close substitutes for bank deposits in terms of liquidity and therefore, much like bank deposits, were likely to be related to spending decisions. However, as monetary analysis is based on the consolidated balance sheet of the MFI sector, the concept of balance sheet identity dictates that a corresponding counterpart must be entered for each newly added monetary variable. So if the Eurosystem adds the liabilities side of money market funds’ balance sheets – that is, the issued money market fund shares – to the MFI sector’s consolidated balance sheet, it follows that the assets side of money market funds’ balance sheets likewise needs to be included. This was relatively straightforward for money market funds because they could be subjected to reporting requirements and their business activities essentially confined to receiving and investing fund assets (the latter primarily in short- term near-bank investments). [...]

The increase in significance of the shadow banking system goes hand in hand with greater interaction between shadow banks and commercial banks. While an unconsolidated analysis of payment flows between the individual financial sector players is of interest from a financial stability perspective, monetary analysis seeks to adjust the monetary and credit aggregates affected by the interaction, if need be, so as to best capture the changes relevant to price developments.

Repo and reverse repo transactions

Secured money market transactions known as repo and reverse repo transactions, which commercial banks conclude with central counterparties (such as Eurex Clearing AG, which until recently was statistically classified as a non- bank financial intermediary), are examples of this. Repo transactions (reverse repo transactions transactions) were originally presented in the banking statistics – which are crucial for monetary analysis – as an outflow (inflow) of funds for the money- holding sector. Yet for the most part, these transactions inherently constitute secured money market transactions between commercial banks in which the central counterparty merely acts as a go-between. Consequently, the money supply is not expanded at the macroeconomic level. These secured transactions grew steadily in importance over the course of the financial and economic crisis owing to the high level of uncertainty in the interbank market. They ultimately had such a significant impact on short-term monetary and credit developments, both in terms of quantity and their month- on- month volatility (see the chart on page 27), that the Eurosystem decided to exclude them from the calculation of M3 and its counterparts in August 2012.

enter image description here

I don't know how the Fed deals with this. Like I said in a comment, the M-terms don't have universal definitions, when it comes to such details.

Note that the ECB has not excluded all the shadow banking from their M3 (not even all the repos), just the repos with "central counterparties".

enter image description here

  • $\begingroup$ Hey, thank you for such a thorough answer. However, I don't understand the line, "Consequently, the money supply is not expanded at the macroeconomic level." -- How can that be? The lender is left with a deposit acct that affects their spending. And the borrower has actual cash. There is a new, cash-like liability being created? $\endgroup$ – Davis Clute Apr 3 '20 at 17:40
  • $\begingroup$ What on earth does this have to do with the question? $\endgroup$ – Frank Apr 3 '20 at 20:00
  • $\begingroup$ @Frank: it has to do with the context in which repos are [or aren't] included in the monetary indicators. I thought that what the questions is about. Of course, the question is vague as it doesn't define "transaction-useful money". $\endgroup$ – Fizz Apr 3 '20 at 20:10
  • 1
    $\begingroup$ @DavisClute: I think they mean that it depends who's doing repo with whom. The ECB decided that banks doing repos with each other (via some third party intermediary) doesn't inflate the money supply (as seen on mainstreet), but repos by others "still count" towards M3. $\endgroup$ – Fizz Apr 4 '20 at 1:23
  • 1
    $\begingroup$ @Fizz Prima facie that sounds like an untenable argument (for a variety of reasons)...but this is exactly what I was looking for with my original question - so thank you :) -- To hit intellectual "rock bottom" on arguments about the endogeneity of money, especially vis-a-vis the shadow banking "system." $\endgroup$ – Davis Clute Apr 4 '20 at 1:27

A repurchase agreement (“repo”) is used by investors to effectively borrow against their portfolio, which can then be used to finance other positions.

  • For the “borrower,” it is the economic equivalent of a loan. A loan is a liability, and money is an asset.
  • For the “lender” it is a short-term asset. It can be an asset in a money market fund, and money market funds are considered part of broad money supply measures. (In some jurisdictions, you can write a cheque against such a fund, I believe. Not possible where I live.) As such, has a relationship to money.

It should be noted that repos are included in M3 in the euro area, so they literally increase M3. Link to ECB website.

  • $\begingroup$ Hey, I understand repo - a close friend worked on Repo desk at JPM. My question is: Is there net "new" money being produced by a repo agreement? It bears striking resemblance to normal deposit banking at Commercial bank. If you put your money into a Commercial bank, the "new" money being produced is the deposit account. I.e. M1 just increased. $\endgroup$ – Davis Clute Apr 2 '20 at 22:04
  • 1
    $\begingroup$ It does create money. People on one end of the chain received money in the form of mortgage loans. People on the other end hold money in the form of demandable shares in money market funds. It creates money just as the lending out of deposits does. $\endgroup$ – dismalscience Apr 3 '20 at 12:10
  • 2
    $\begingroup$ Repo “converts” a long-term bond (not considered money), into a money market instrument that can be used to back a money market fund (“broad money”). All else left equal, the broad money supply goes up. $\endgroup$ – Brian Romanchuk Apr 3 '20 at 14:03
  • 2
    $\begingroup$ @Frank - if the transaction raises M3, it “creates money”. It doesn’t create M1, but there’s a reason why broad monetary aggregates exist. $\endgroup$ – Brian Romanchuk Apr 3 '20 at 21:14
  • 1
    $\begingroup$ @Fizz: money market fund liabilities are in M3 (in all countries, I believe). Repos are on the asset side of their balance sheets. Even if you don’t want to directly include repos, they show up indirectly. $\endgroup$ – Brian Romanchuk Apr 3 '20 at 21:21

I will add this as answer:

Your comment "If you put your money into a Commercial bank, the "new" money being produced is the deposit account."

That's not how banks increase the money supply. They increase the money supply when they issue a loan. They do not require reserves to do this. Even in times when reserve requirement is 10%, the banks can create, say a million dollars in new deposit accounts and then put 10% of those in the central bank. More accurately the rate of interbank spending is so high that by the time they have to settle with the central bank what gets parked there is from a mix of sources. They can park securities instead as reserve collateral, I belive. This is how it works and I have a number of links to prove it. Here is rich source of references https://www.debtdeflation.com/blogs/2009/01/31/therovingcavaliersofcredit/ This myth that fractional reserve banking operations work by creating loans based on existing reserves, which are spent at another bank to create new deposits to create new loans, turning 10k into a million, is just that : a myth. Banks can issue as much new money as they deem healthy.

Someone putting 10Eur in a bank and getting deposit money in return isn't increasing the money supply. The bank would not spend its reserves, it would invest them, but today all banks need do is issue loans.

Repo is just a swap with an agreed future buyback. I don't see an analogy with the above debt creation process.

  • $\begingroup$ "as much new money as they deem healthy" and as capital requirement regulations allow them: economics.stackexchange.com/questions/27660/… $\endgroup$ – Fizz Apr 3 '20 at 5:43
  • $\begingroup$ @Fizz yep...... $\endgroup$ – Frank Apr 3 '20 at 6:04
  • $\begingroup$ @Frank I feel like you & I are going to debate this ad naseum. I see your point, but pedagogically I think it gives the wrong intuition - it leaves people thinking there is so mystical process going on. I get that banks can create M1 (deposit accts) as much as they want...I get that... $\endgroup$ – Davis Clute Apr 3 '20 at 17:45
  • $\begingroup$ @DavisClute in fact I really think you don't get it . How the hell does giving a bank your cash create money? Is that really what you think? $\endgroup$ – Frank Apr 3 '20 at 20:56
  • $\begingroup$ @Frank I think you should take a step back to cool down. I come here for fun, light-hearted intellectual banter. Not vitrol. $\endgroup$ – Davis Clute Apr 4 '20 at 1:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.