NYSE Margin Debt Just Shy of New All-Time High in July

The latest reading on NYSE margin debt was just shy of an all-time high.  Borrowing to execute securities trades has not waned much at all in recent months as the July data showed borrowing of $382.1B which was slightly below the all-time record of $384.3B seen in April of this year.  The July reading is up from $376B in June.  Clearly, there’s still a lot of confidence in the minds of traders who are borrowing to execute trades as NYSE margin debt balances certainly appear to have a long-only bias:



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:

  • Anonymous

    EDIT: ” Clearly, there’s still a lot of confidence in the minds of traders who are borrowing to execute trades.”

    SHOULD READ: ” Clearly, there WAS still a lot of confidence in the minds of traders who are borrowing to execute trades.”

  • Mr. P

    No problem….

    Take lots of risk, the Chairman and the Fed have your back!

  • http://highgreely.com John Daschbach

    There are lots of other ways to think about this data. I think this is an example of not thinking critically about the data and reaching a conclusion fits pre-concieved views.

    1.) The margin debt roughly correlates with the S&P 500 level. Does that tell us anything? Just that some part of the market operates on margin.

    2.) Margin debt vs S&P 500 is 0.25 and in 1994 it was 0.14 so margin debt to S&P ratio has increased. Does this tell us that there is more leverage in the system today?

    3.) Margin debt yields were about 2.3 times higher in 1994 and so the expected return on equities was higher. The relative cash flow associated with servicing margin debt was 1.3 times higher in 1994.

    4.) NYSE margin debt to total market cap (using DDDM01USA156NWDB and GDP [nominal]) is 2% today and 8% in 1994.

    I would argue that the best way to look at this is that some portion of the market operates on margin as it always has but it is more risk adverse than in 1994. One could look at other dates, but this seems like one of those low value pieces of data.

  • Andrea Malagoli

    Here is one manifestation of the Fed’s transmission mechanism to inflate asset prices…. get very cheap money and leverage a market that is prevented from correcting.

  • Andrea Malagoli

    … until it does …

  • Notagain

    Just watch price and large caps and they don’t look healthy here .

  • inDC

    New all-time highs on margin debt ALWAYS = stock crash within months

    It is one of the few untainted signals we have left in this whole corrupted market

    Good luck to all

  • SS

    O/T, but did you see this article? It seems to be targeted at MR. Any thoughts on it?


  • http://orcamgroup.com Cullen Roche

    Yes, was just discussing this on Twitter with some folks. The author says:

    “(banks) make money by charging interest on lending, in more or less the same way any of us could make money by charging interest in lending. Banks just do it on a much larger scale.”

    This is categorically false. When banks lend they increase purchasing power. When you lend you do not. So the comparison is totally false and renders the article largely misleading. Furthermore, the author implies that the aggregate banking system has a constraint that is completely different than the govt’s constraint. It implies that the govt is not constrained by pvt output, which it is. The aggregate banking system is also constrained by output. So it’s a fallacy of composition. We all know that banks can issue liabilities well in excess of the owners’ ability to pay them back. Just like the govt can issue more liabilities than output can digest. Slightly different constraints, but both related to pvt output. So the fallacy of composition falls apart due to a micro extrapolation to the macro. Banks, in the aggregate are constrained by capital and output just like the govt is. Neither one has a Widows Cruse because neither one can fabricate output from thin air. And one could argue for ages who is more capable of producing viable output, banks or govt money issued money. But the point is, both are ultimately constrained by output and not the ability to issue money. In this sense, the aggregate banking system is pari passu with govt by virtue of having been outsourced money creation by the govt.

  • Anon

    It doesn’t just “roughly” correlate, the correlation is well above 0.9 …

    Secondly, if you look at the data carefully you see two additional details:

    – very strong long bias (going short on margin is almost invisible)
    – margin debt often leads the S&P right before it peaks

    The conclusion is obvious: speculative, dumb money flows probably fuel the “crazy” portion of speculative runups, unintentionally fuelling serious correction via a short squeeze. If you think the macro picture is a boom then the next dip buying opportunity would be once a good portion of those leveraged positions liquidated. If you think it’s a bust then it’s an arguably suitable moment to be short.

    Unless this time is different, right? ;-)

  • Anon

    You are so confused.

    By your argument, looking at the data, the previous peak, 2000, was very similar.

    Except that in 2000 interest rates were at over 5%, for almost a decade:


    And in the 1950s, despite a near zero Fed Funds Rate, there was no boom and bust.

    What matters to speculation are two things:

    – financial deregulation (1928 again baby!)
    – a positive Zeitgeist that gets even the last groups of buyers into the market, those who missed the run up and feel left out.

    Interest rates, not so much. But nice anti Fed comment in any case.

  • http://orcamgroup.com Cullen Roche

    They really just miss the point of endogenous money though. It’s not about the idea that the govt’s got the ability to tax and therefore make banks whole at the micro level. It’s that bank money dominates our everyday transactions. Bank money is what influences prices, employment and output at the transactional level. It’s not about govt money. Govt money is secondary in all regards. When you buy goods at the market the bank doesn’t make the purchase for you. You decide on a price and transaction and tell your bank the price to settle your payment at. Your bank is a middleman and the Fed is a middleman for the middlemen. MMT builds this story that implies the govt is the center of the transaction and without their money we’d all be wallowing in misery trying to figure out how the economy works. Yes, govt money matters. Yes, govt money facilitates and supports inside money. I don’t deny that. But govt money doesn’t rule the monetary roost. It is secondary to bank money in all real economic transactions. And that’s why endogenous money matters. It is the money that matters at the point of transaction and serves as the dominant form of money in determining real economic outcomes, prices and wages. MMT just creates their own little version of the money multiplier and neoclassical thinking by focusing on reserves. It’s all backwards. It’s just a reworked version of the same flawed govt centric version of the system that neoclassicals use….

  • Geoff

    Cullen, you made two interesting points above. The first is that banks are constrained by output. We talk a lot about how banks are constrained by capital but not as much about output. But it makes total sense. If borrowers aren’t making productive use of the money, then defaults will be a huge problem down the road. :)

    Your second point about the banking system ranking equal to govt is obviously more controversial. Those schools of thought who use a “hierarchy of money”, with govt money at the top, are going to disagree. They seem to feel that the conversion feature is somehow important. The fact that deposits must be converted to govt money on demand, and that banks ultimately settle among themselves with reserves, somehow means govt money ranks higher.

    But even if one were willing to grant them this point, it doesn’t mean that govt money “rules the roost” as you say. Even if govt money ranks higher, it doesn’t mean that it plays anything more than a facilitating role. And it certainly does mean that there is some sort of base money multiplier effect.

    Who cares if banks use reserves to ultimately settle among themselves? What matters is what money we all use on a daily basis, and that is inside money for the most part.

    But I must confess that I just sold another one of my gold coins and now have $1400 of govt money burning a hole in my pocket. :)

    A large part of which is in $1 bill denominations (not to give away where I’m going tonight) :)

  • aaaxn

    “When banks lend they increase purchasing power. When you lend you do not. ”

    Is it really true? Suppose Alice is really credible person and there is secondary market for her IOUs. When Alice borrow $10000 from Bob she has the money and can spend it. But Bob still have her IOU which he can sell on secondary market so he still have his buying power (at least part of it). Money supply increased and so did purchasing power. Not that different from how banks operate.

  • SS

    His point on seiniorage is also stupid. Not only is his example wrong, but he’s implying that the government makes money from creating mo ey which contradicts MMT.

  • SS

    If all of us were “really credible” then it might be valid, but we’re not.

  • Johnny Evers

    Purchasing power always increases when you make any kind of loan. Loans or deficit spending move future spending into the present.
    However, if spending does not create future growth, then the effect is temporary.
    Most economic formulas are fixated on the here and now. Make a loan, create money, it stimulates the economy. There doesn’t seem to be a way to express in the models and the formulas if this is sustainable; as a result, it’s put out of mind, or, worse, we become fixated on today, so that permanant ‘stimulus’ (either private or government borrowing) becomes the policy.

  • http://highgreely.com John Daschbach

    But Paul who bought the IOU from Bob has lost whatever MZM money he used to buy the IOU with. Completely different from how banks operate. Paul and Bob have done an asset swap. They had to have the money before they lent it out or bought the IOU. The can create the loan with no money (reserves). If the money flows to other banks, the originating bank goes out and finds the reserves. That’s what the Fed Funds market is. Overnight loans between banks to meet reserve or liquidity needs.

    If Bob has a line of credit from a bank, he can use that to offset the loss of money in the IOU. So, yes, money supply can increase via this process, but it has to show up in bank balance sheets.

  • http://highgreely.com John Daschbach

    If it had a correlation of 1 it tells you either that the margin debt level is the only independent variable that impacts the market, or that it is just another way of measuring the dependent variable. The high correlation and small share of the market it represents shows that it’s mostly a dependent variable of the S&P 500. It tells you hardly anything.

  • aaaxn

    It was not asset swap, because IOU did not exist prior to transaction. Sum of assets increased.
    And transaction could be done without involvement of base money. They could just swap IOUs (Alice gets Bob IOU and Bob gets Alice IOU)
    Later Bob can go and sell Alice IOU on secondary market.

  • aaaxn

    So you agree it is only technical difference and not fundamental one.

  • http://highgreely.com John Daschbach

    No, you need to learn a little double entry book keeping. Here is a trivial example for this situation.

    It is absolutely an asset swap!


    Bob: asset: $100 cash
    liability: $0
    Alice: asset: $0 cash
    liability: $0
    Sum: $100


    Bob: asset: $100 IOU
    liability: $0
    Alice: asset: $100 cash
    liability: -$100 IOU
    Sum: $100

    Sum of assets Before – Sum of assets After = 0

    Those are the facts. It’s 5th grade math. My kids have had to do this in school.

  • aaaxn

    Where did I say that assets + liabilities increases? I said that assets grows.
    Look at your example. Sum of assets before transaction = $100.
    Sum of assets after transaction = $200.
    Asset swap does not do that.

  • Andrea Malagoli

    We’ll see what happens the next time the Fed barely mentions the vague idea of tapering … It is not the absolute level of interest rates that matters, it is the relationship to the perceived equity risk premium, which is now clearly quite lower than it was in 2000.

    Both of your bullets cannot be substantiated by the data just as much,if you talk about being confused. Volumes have been very low, and there is no demonstrable relationship between flows of money “into” the market and market trends.

  • Qcl

    Increase in an asset has to be matched either by decrease off the same amount in Medium of Exchange of another asset or equal increase in liabilities or equity . It is basic BS accounting .

    Think of your 100 dollar bill , it had to come from somewhere. You could have earned it , stolen it , found it etc. but it had to be , in this case literally printed . Otherwise it would be counterfitting . If Alice was given 100 $ asset it must have been matched by equal 100 $ liability. Only other way is financing from equity ( Alice has none and that’s why she borrows) or counterfitting .

  • http://asymptosis.com Steve Roth

    Seems like margin debt is a trailing indicator on the way up, and a leading indicator on the way down.


  • jt26

    What’s more surprising is margin debt is not higher. As a retail investor, you can actually borrow at 1.6% and do a large cap dividend carry, on say DVY @ 3.4%. Yikes, how often does that happen. (Of course, part of that is margin rates are so much more competitive than 20 years ago.). Scary stuff.

  • nameiusedlasttime…again

    “Think of your 100 dollar bill , it had to come from somewhere. You could have earned it, stolen it , found it etc. but it had to be, in this case literally printed . Otherwise it would be counterfitting . If Alice was given 100 $ asset it must have been matched by equal 100 $ liability.”

    Yeah, imagine I just pretended I had $100 and you had confidence in that assertion. Theres $100 liability right there backed by nothing but my mouth.

    Impossible, right?

  • http://highgreely.com John Daschbach

    If a fraction of the market employs a consistent amount of leverage then some would confuse this with leading on the way up and trailing on the way down. It’s just the natural manifestation of hysteresis in the margin process.

    People like Keen lack critical scientific thinking skills. He wonders if margin debt causes the market to rise, or if a rising market causes more people to take on margin debt. Neither to any significant extent is the correct answer. Margin debt is a small fraction of the market (2% today). If people think 2% drives the market then the market model is worthless. A rising market doesn’t necessarily cause more people to take on margin debt, but it leads to an expansion of this debt by people who employ this strategy.

    Keen makes another incorrect analysis, comparing the stock market to the CPI. This ignores the very basic rational of investing in the stock market, which is to capture a share of future profits. In a world with zero inflation, but 2.3% annual growth, the DJIA would have grown by his factor of 10 in 100 years, exactly as it has in constant dollars (and below the mean compounded growth rate over 100 years).

    Which tells us that if the market represented a constant share of GDP (it doesn’t, it’s grown far higher over 100 years) it’s priced about correctly. If you only run the data back to 1947 when many of these numbers get more reliable the current market is also at 50% of it’s high and low. If we had the numbers for corporate share of the economy it would be lower, probably between 25% and 40% of it’s highs (1962, 1990). Of course it can easily go back down to it’s 1983 low, but it’s clearly not exorbitantly high relative to history (even without considering the increasing role of public companies in the economy).

  • Qcl

    My confidence will perish as soon as payment is due. Situation you described is just another way of counterfeiting .

  • Notagain

    if you have looked at enough indicators you will eventually notice they are either correlated to price or the r.o.c of price .

  • http://www.fbs.com/ Victor Wooten

    Don’t keep a dog and bark yourself.

  • http://www.justaluckyfool.wordpress.com Justaluckyfool

    Quote Cullen Roche, ” When banks lend they increase purchasing power. When you lend you do not. So the comparison is totally false and renders the article largely misleading.”
    …May I ask ? Isn’t the comparision false because of as Frederick Soddy states, (Paraphrase) There is two types of lending; Real or Genuine lending (that which people do when they must surrender all rights to the capital that is lent and they must be owners of that capital) and ‘Fictitious lending’ (that which allows bank to print money they do not own let alone not even have the rights to surrender its ‘good faith guarantee to conversion albeit if they print enought the Fed will ‘have to cover it so there is no ‘systemic failure’).

    Quote Cullen Roche, “But the point is, both are ultimately constrained by output and not the ability to issue money.”
    …May I ask ? Isn’t as a RM thought that The Fed can print currency ‘in unlimited amounts’ constraint is only that of self-imposed rules. Banks will be limited by Fed imposed ‘restraints’ (like 6% capital requirement) but this is only an after the fact restraint that is,it can only be applied after the banks over print. May I ask ? Isn’t the ‘constrained by output’ really only another self-imposed restraint ? For surely at 0% interest for 10 years would tax the printing press to exhausion.

    Please, a profound answer to some foolish questions would be appreciated. I have a really lousy writing skill and communication skill so hoping that you can get pass that knowing that I sincerely believe : Prag. Cap is providing some of the best information ‘on the net’.