Wednesday, October 20, 2010

Fed's Lockhart: Let's Print a Trillion Dollars Plus

"If we're going to pursue another round of quantitative easing, it has to be a large enough number to make a difference," Atlanta Fed President Dennis Lockhart said in an interview on CNBC television. He then stated he was thinking $100 billion a month."As a monthly number ($100 billion) is fairly consistent with what we did before, and so I think it would certainly be in the range of numbers one might consider ... but if you were talking about $100 billion as simply the overall program, I think that's too small," he said

Thus, over a 12 month period, we are talking $1.2 trillion.

It is important to keep in mind that $1.2 trillion is what is known as super money. The actual amount of money that could find its way into the economy could be a multiple of the $1.2 trillion super money.

$3.6 trillion (a multiple of 3) is not out of the question. A trillion is very inflationary, 3.6 trilllion takes it to an entire new level.

17 comments:

  1. Wenzel,

    For the "OMG hyperinflation is here!"-confused amongst us, can you please briefly explain how that $1.2T flows into the economy?

    So, the Fed prints $100B a month, and it uses it to buy _______ and then the sellers of ______ use the money to..., etc.

    Thank you

    ReplyDelete
  2. I don't understand. Was he talking about the base? If he's talking about mzm, what he's saying makes sense. If he's talking about the base, then I must be looking at Fred data wrong. Increasing the base by trillions in the next year? They already doubled it and they still haven't drained that first round of large QE they began when everything tanked in 2008. It's going to snowball. Maybe I'm missing something? Please tell me if I've got it totally wrong because this news kind of creeps me out.

    ReplyDelete
  3. Taylor Conant - see www.mises.org/literature/mysteryofbanking.pdf , chapter 9 and following. In a nutshell the FED purchases an asset; the money to buy it is created out of thin air; this becomes an increase in the reserves of a commercial bank when it deposited by seller in his bank; Commercial banks may pyramid on top of this by extending loans *beyond* their reserves (the subsequent round of money creation); The money issued by the first bank is deposited into others. The cycle repeats - more loans issued and more money created. This puts much more money into circulation than the amount of the FED purchase, hence the term money multiplier.

    ReplyDelete
  4. Hi Anon,

    I am familiar with Rothbard's work. I appreciate the link but that is not the question I was asking.

    The question I was asking is, "The last injection of $1.5T did not result in hyperinflation, while will this one?"

    I am not so silly or stupid as to believe that injection did "nothing" or that it was "harmless". However, it does seem like it has not (yet) created hyperinflation. As Wenzel has noted, repeatedly, this is likely in large part due to the fact that they have been paying interest on excess reserves.

    Are they not going to pay interest on excess reserves? Will they, instead, even charge a fee for maintaining excess reserves, spurring banks to lend (as Mankiw has been having wet dreams over)?

    I want Wenzel to explain the step-by-step hypothetical for how this money is going to find its way into the economy instead of into banks' excess reserve piles, when the last go around did not. Then, we can have a discussion and critique it, if necessary.

    Until then, it's coming across as a lot of hyperventilating and a lot of naive taking of the Fed at their word.

    I understand the danger is looming and this thing could get ugly easily, without any warning or desire on the part of the Fed or other interested authorities. What I want to know is HOW, specifically?

    SHOW... ME.... THE MONAAAAAYYYYY!!

    ReplyDelete
  5. @Taylor Conant

    I'm sure I have wriiten this before,but here goes:

    QE 1 was the Fed buying MBS junk directly from Goldman Sachs, Morgan Stanley etc. They, for a variety of reasons, put the money into excess reserves.

    QE2 is not going directly to GS and MS, it will be, via the Primary Dealers, be bought from the public, everyone from pension funds to farmers, if they are selling they have plans for the funds. It's not going to end up on the first pass in excess reserves--later passes it's iffy after it is deposited, but I don't think it will be put in excess reserves at that point either. But the key is the first round, that will be out in the system--no question. Other rounds will determine how big a multiplier we get, but a trillion (if that's what they print will cause enough trouble)

    ReplyDelete
  6. Wenzel,

    I know you've written about this but I don't think you've gone into much detail aside from general projections about what might happen. So, if you don't mind, keep walking me through this. Next questions:

    1.) Who are all the Primary Dealers who will be transacting this stimulus, and why are they different from the first set last time?
    2.) Do the primary dealers not have their own holdings of Treasuries they could sell to the Fed instead of selling their clients Treasuries?
    3.) Say the money somehow winds up with the Primary Dealers, after they have sold something to the Fed... what might they do with the money?

    ReplyDelete
  7. 1. The NY Fed has a list of thre primary dealers.Almost all Fed transactions use the primary dealers.

    QE1 did not, the Fed wentd directly to the big banks they pumped money into.

    2. Primaries have some, but nowhere near enough.

    3. They can do anything they want with it, including loaning it out or putting it in excess reserves, but they don't have anywhere near the securities on their own books to do it themselves. It won't happen this way.

    ReplyDelete
  8. PD's were and are the counterparties in all OMOs, whether Treas/Agency/MBS. The set has changed by just a few members in the last two years. The $1.25 T in MBS purchases were covering bad bets in what had become a highly illiquid mkt. Banks were trying to hide the losses in several ways. Accounting rules were changed to accommodate as well. The Fed's purchases directly covered some of the bad bets but also lifted the whole MBS market. I believe most banks were content to simply leave any MBS funds as excess reserves, but I don't have smoking gun evidence of this. However, Treasury OMOs are a different animal. Consider that they have always been about influencing short term interest rates through direct manipulation of the monetary base. Sustained operations, esp. permanent ops as opposed to temporary ops, have historically influenced M2. The chart I posted of the sales in 2008 is but one example. The PD's maintain large inventories of unhedged Treas paper, and they also act on behalf of their clients. What will the PD's do with the money? For transactions in which they acted as a dealer, they pass the funds on to the clients--banks, cb's, the public, etc. For transactions in which they acted a principal, they can do whatever they want. In a post-IOER world, how do we know they won't simply park funds at the Fed? They've been conditioned for at least a decade to adjust their short term lending according to Treas OMOs--that is, to take the funds and directly lend them in the markets. But even if that does not occur, let's look at the numbers. Total Treasury and Agency (they aren't reported separately) of all US banks total only $550 B. The banks will acquire more Treas paper to sell to the Fed over the coming year, but they also won't sell their entire stock. So let's say they sell $500 B (and I believe this is high) to the Fed and keep 100% of the funds there to earn IOER. For a $1 T QE2 program, that leaves $500 B that will flood the money supply, which will be multiplied and will be inflationary. Let's not forget that QE Lite will itself add another $320-$360 B over the next year. The minimum for QE2 is $500 B, so we're talking about a min total of $820-$860 B. The numbers are astronomical and the banks simply don't have the supply for these ops to be sterile.

    ReplyDelete
  9. PD's were and are the counterparties in all OMOs, whether Treas/Agency/MBS. The set has changed by just a few members in the last two years. The $1.25 T in MBS purchases were covering bad bets in what had become a highly illiquid mkt. Banks were trying to hide the losses in several ways. Accounting rules were changed to accommodate as well. The Fed's purchases directly covered some of the bad bets but also lifted the whole MBS market. I believe most banks were content to simply leave any MBS funds as excess reserves, but I don't have smoking gun evidence of this. However, Treasury OMOs are a different animal. Consider that they have always been about influencing short term interest rates through direct manipulation of the monetary base. Sustained operations, esp. permanent ops as opposed to temporary ops, have historically influenced M2. The chart I posted of the sales in 2008 is but one example. The PD's maintain large inventories of unhedged Treas paper, and they also act on behalf of their clients. What will the PD's do with the money? For transactions in which they acted as a dealer, they pass the funds on to the clients--banks, cb's, the public, etc. For transactions in which they acted a principal, they can do whatever they want. In a post-IOER world, how do we know they won't simply park funds at the Fed? They've been conditioned for at least a decade to adjust their short term lending according to Treas OMOs--that is, to take the funds and directly lend them in the markets. But even if that does not occur, let's look at the numbers. Total Treasury and Agency (they aren't reported separately) of all US banks total only $550 B. The banks will acquire more Treas paper to sell to the Fed over the coming year, but they also won't sell their entire stock. So let's say they sell $500 B (and I believe this is high) to the Fed and keep 100% of the funds there to earn IOER. For a $1 T QE2 program, that leaves $500 B that will flood the money supply, which will be multiplied and will be inflationary. Let's not forget that QE Lite will itself add another $320-$360 B over the next year. The minimum for QE2 is $500 B, so we're talking about a min total of $820-$860 B. The numbers are astronomical and the banks simply don't have the supply for these ops to be sterile.

    ReplyDelete
  10. "For transactions in which they acted as a dealer, they pass the funds on to the clients"

    Above, replace "dealer" with "broker"

    ReplyDelete
  11. Gentlemen,

    What is the chance that the Fed actually ends up buying more fraudulent MBS from the PDs with QE2, and that money gets all or mostly parked as excess reserves?

    ReplyDelete
  12. @Taylor Conant

    The Goldman, Morgan Stanley et al. That's done. The Fed is going to buy Treasury securities.

    @Bob English

    It's mostly a technical point that doesn't have much to do with analysis, but I don't believe MBS was done via PDs the way they do Treasury dealer work. MBS was sold by the banks, who happen to be PDs, but it probably wasn't even the same desk, and most certainly it was in house paper, not paper bought in the market as agency transactions.

    ReplyDelete
  13. Wenzel,

    So there is no chance that all this QE2 talk was not an elaborate ploy to prepare the public with an excuse for the Fed to purchase more MBS, given that new problems in the MBS market are coming to light at the exact same time, all but necessitating another round of weightlifting on the part of the Fed to get that crap off their balance sheets before their equity takes a big hit?

    ReplyDelete
  14. It's a different kind of problem. B of A has huge exposure to foreclosuregate, but that's not because they are necessarily holding the paper, it's because they serviced the loans, and are responsible if there is bad paper work--even after the paper is out of their hands.

    ReplyDelete
  15. @Wenzel: Yes, it's a technical point and I understand the source of confusion. FRBNY hired outside consultants, such as Wellington and Blackrock, to aid with execution of the ops initially. In March 2010, FRBNY took over completely and did all the dollar/coupon rolls over the summer. However, the counterparties in the ops were always the FRBNY and the PD's. No banks that were not PD's participated directly in MBS POMO--they would have had to have used PDs as agents. Now, JPM, Goldman and others were compensated in other ways outside of MBS purchases, such as through payments to AIG and by sheltering toxic assets in the Maiden Lane vehicles.

    @Taylor: Though I don't think it's likely in the near term, if the Fed does decide to make QE2 mainly about MBS, the banks have $1 trillion on hand, and I would not expect it to be inflationary for the same reasons as outlined previously. The securitization flow could get pretty jammed in any number of ways and there is risk of the MBS market seizing up if certain big players aren't there to participate (including the GSEs themselves), so it's something to be on alert for.

    ReplyDelete
  16. @English

    I think it is misleading if you think of it as a PD operation. As you point out no non-PDs were invloved. This was not a normal PD operation, where they acted as dealers. It was fully to bail them out. Further, it is possible some PDs did not participate. Thus it was not done with PDs perse but a PD subset.

    The confusion comes from thinking of it as a PD operation.

    ReplyDelete
  17. I've consistently said the MBS purchases were a different animal, especially as to effect. Here's another nugget: they were not conducted through auctions, but negotiated directly with the PD counterparties based on prevailing market conditions. As to how many PD's participated, we can only speculate, but the size of the program suggests it was broad based.

    On the day the MBS program was announced, the MBS stock of the largest 25 domestic banks (of which PDs are a subset except for a few foreign banks) was $682 B. As total Fed purchases were $1,250 B, there would have been significant agency transactions.

    ReplyDelete