With There Ever Be Another "Tight Money" Recession?

[stone]

MG, can you explain HOW the fed would raise rates much above inflation when the amount of bank reserves flooding the system is as large as it is now?

To my mind the fed has painted itself into a corner. They just don't have the means to raise rates much above inflation even if they wanted to.

You can debate all day about whether they would want to but that is all by the by if they don't operationally have that option when the amount of bank reserves flooding the system is beyond what they are capable of reeling in.

Perhaps the consequence of having a large government debt is that the Fed looses its capacity to induce above inflation interest rates whilst there is a trade deficit. Japan manages to have above inflation interest rates (despite trying to do the opposite) because they have a trade surplus that is strengthening their currency despite low nominal rates and so creating slightly positive real rates.

Above inflation interest rates are a way to strengthen the exchange rate and so make imports cheaper. You don't need "capital inflows" for people to be employed unless you are meaning that lower commodity costs help industry and employment. Remember banks can and do "with the stroke of a pen", create as much money as they think can be profitably used.
I guess "capital inflows" fueled the housing bubble and that created jobs building and selling unneeded housing but I don't think the long term effects of that were positive. To my mind the economy works much better when exchange rates reflect trade balances rather than "hot money" capital flows. The ideal is for the real economy to be able to make real investment in the productivity of what is actually needed. That avoids bubbles and waste.

[moda0306]

MG,

You could still be right... I'm just saying that in spite of any reason he may have said he raised rates at the time, it seemed like it was pretty par for the course fed policy.

Glad we're seeing eye to eye for a bit... fun while it lasted .

[MachineGhost]

You could still be right... I'm just saying that in spite of any reason he may have said he raised rates at the time, it seemed like it was pretty par for the course fed policy.

Glad we're seeing eye to eye for a bit... fun while it lasted .

I re-checked today and the reason why Greenspan raised rates in 1993/1994 is because the economy was indeed expanding but he wanted to head off inflation before it arrived by raising from 3% to 3.25%.  He believed that by doing so it would lower inflation expectations and hence lower long term interest rates and produce a "soft landing".  But what happened was people paniced because an increase in interest rates by the Fed always started a series of increases historically, and combined with the high levereage of hedge funds, it was panic selling galore.

So I reiterate it won't take "breaking the back of inflation" to raise real interest rates and take the PP out to the woodshed.  

MG

[MachineGhost]

MG, can you explain HOW the fed would raise rates much above inflation when the amount of bank reserves flooding the system is as large as it is now?

The Fed raises or lowers the Fed Funds Rate by reducing or expanding the bank reserves.  The Fed controls three things: Discount Rate (lender of last resort), Fed Funds Rate (bank reserves) and reserve and margin requirements.

You can debate all day about whether they would want to but that is all by the by if they don't operationally have that option when the amount of bank reserves flooding the system is beyond what they are capable of reeling in.

I don't understand why you feel this way.  If the Fed can increase the bank reserves, why can it not take it all back?  All the reserves aren't sloshing around the system anyhow, they're parked at the bank's checking accounts with the Fed.  Monetary velocity is low.

MG

[moda0306]

We have massive currency inflation right now.  Price inflation is to my mind inevitable.  The only question is when. 

Do we really have currency inflation if bonds are just about as much a currency as cash reserves are?  Do people invest in treasury bonds because of the measly 0-1% interest they pay?  Do they invest in treasury bonds because they think the treasury will borrow or tax to pay them back, pulling money from somewhere else in the economy? Or do people invest in treasury bonds because they wish to save in the common medium of exchange, and .2% interest is better than 0%?

The third option drives the demand for treasury bonds, and the first two are far secondary.

There is a theory out there that the Fed could QE all of the national debt, and we still wouldn't have much inflation... I'm 80% confident they are right.  The bond markets wouldn't know what to do without a treasury rate peg, but that might even put people in "rush to safety mode" and we might be surprised with the resulting will to hold dollars if people can't price bonds effectively.  Peoples' balance sheets, after such an event, would look almost identical to before.

[MediumTex]

Does anyone think that the Fed would ever do something like this again?  I tend to think that it wouldn't.  I'm actually sort of surprised that it ever even happened once.

Very easily.  Greenspan did it in 1993/1994.  All it will take is a heterodox theory and enough proponents of such on the Board. 

I don't recall there being a recession in 1993-1994.

[moda0306]

MT,

Look at the comments since that one... we've established that MG was misremembering and Greenspan simply wanted to hoard off potential future inflation now that a recovery was in place.

[Wonk]

Does anyone think that the Fed would ever do something like this again?  I tend to think that it wouldn't.  I'm actually sort of surprised that it ever even happened once.

It takes brass balls to raise rates to 20%, so yeah, it's unlikely, but ultimately possible.  All we would need to induce another tight money recession would be to raise rates to 4% right now and keep them north of price inflation.  Gold and equities would go south in a hurry.  But there's exactly 0% chance that will happen, so we can worry about that later...

[stone]

MG, can you explain HOW the fed would raise rates much above inflation when the amount of bank reserves flooding the system is as large as it is now?

The Fed raises or lowers the Fed Funds Rate by reducing or expanding the bank reserves.  The Fed controls three things: Discount Rate (lender of last resort), Fed Funds Rate (bank reserves) and reserve and margin requirements.

You can debate all day about whether they would want to but that is all by the by if they don't operationally have that option when the amount of bank reserves flooding the system is beyond what they are capable of reeling in.

I don't understand why you feel this way.  If the Fed can increase the bank reserves, why can it not take it all back?  All the reserves aren't sloshing around the system anyhow, they're parked at the bank's checking accounts with the Fed.  Monetary velocity is low.

I totally agree about the low velocity. Perhaps saying that the system was steeped in excess reserves might been a better term. It is the low velocity excess nature of those reserves that makes them such a dampener on treasury rates. They have little use except for bidding up the price of treasuries.

I don't see what mechanism the Fed has at its disposal to take those bank reserves back. If they tried to sell off their stock of MBS and (since operation twist) longer dated treasuries, that wouldn't reel in anything like the amount of reserves they initially paid for them. The Fed is on a ratchet. They have unlimited means to expand their balance sheet and "inject liquidity" but very limited means to get it back. If you went to a garage sale with the express purpose of being benevolent to your neighbor and spent you life savings on paying wildly over the odds for their junk, you couldn't then expect to get your money back by selling it back to them.

Increasing the rate paid at the discount window would cause individual stressed banks to pay higher rates but certainly wouldn't increase treasury rates when other banks were out there with excess reserves that obviously they wouldn't lend to stressed banks or the stressed banks would never have become stressed and even thought of borrowing from the discount widow. 
The Fed has no source of funding to pay high rates on reserves directly. Perhaps the treasury could "bail out the Fed" but imagine the deficit and political rawkus that would cause.
Reserve requirements don't increase treasury rates. They do make it more expensive for banks to hold customer deposits but that is very different from raising treasury rates. My guess is that if reserve requirements were raised a lot, then banks would start charging customers that wanted to make deposits. That would actually cause negative treasury rates because people would bid up the price of treasuries so as to avoid those charges.

[AgAuMoney]

I don't see what mechanism the Fed has at its disposal to take those bank reserves back. If they tried to sell off their stock of MBS and (since operation twist) longer dated treasuries, that wouldn't reel in anything like the amount of reserves they initially paid for them. The Fed is on a ratchet. They have unlimited means to expand their balance sheet and "inject liquidity" but very limited means to get it back.

What about the minimum reserve requirement (deposits:reserve ratio)?  They could easily raise the reserve requirement so those reserves were no longer excess.

The Fed has no source of funding to pay high rates on reserves directly.

That's not correct.  The Fed has essentially unlimited funding, and they can utilize it however they deem necessary.  The only limit on the Fed is the 6% limit on profit, excess of which is remitted to the Treasury.  Their easiest source of funding is simply creating money from nothing.  It is from this source they made a total of over $16 trillion in emergency loans around the world starting in Dec 2007, as documented in the limited audit performed by the GAO and reported in July 2011 (GAO-11-696, 266 pages).  (Note that Bernanke denies the numbers from the GAO audit and claims the maximum total commitment peaked at $1.5 trillion.  Note that Bernanke was speaking of the peak at one moment in time and the GAO was speaking of the total over time so Bernanke was simply obfuscating the truth.)

Reserve requirements don't increase treasury rates. They do make it more expensive for banks to hold customer deposits but that is very different from raising treasury rates. My guess is that if reserve requirements were raised a lot, then banks would start charging customers that wanted to make deposits.

I don't believe they could do that.  Increasing reserve requirements DOES reduce the creation of money.  The inverse (reducing) has been demonstrated over and over again.  And if less money is being created, liquidity will be reduced.

As for treasury rates, those are set at auction, and currently the Fed is directly setting the rates by bidding enough to consume the entire auction.  If the Fed were to stop bidding, rates would almost certainly increase, and given the Fed is currently buying the majority of the 10 and 30 year treasuries, with them out of the picture the rate increase could be dramatic and many claim there is a real risk of an auction failure.

[stone]

AgAuMoney

  Increasing reserve requirements DOES reduce the creation of money.  The inverse (reducing) has been demonstrated over and over again.  And if less money is being created, liquidity will be reduced.

I totally agree that increasing reserve requirements does reduce the private bank credit creation of money offset by private debt BUT it does not get rid of bank reserves. Unless I'm in a tremendous muddle with all of this, that is an important distinction. Unless bank reserves themselves get reeled in (putting the toopaste back in the tube) "tight money" won't come about.

As far as the Fed making overdrafts to itself to pay interest on bank reserves- is that really plausible? It would be a snowballing scenario. The money paid out as interest would itself get parked at the Fed and would itself "earn" interest. It seems to me to be one thing for the Fed to loan money to RBS with the intention of being paid back and quite another to loan money to itself to pay out as interest when it then has no means of ever closing down that credit creation

[KevD]

You can debate all day about whether they would want to but that is all by the by if they don't operationally have that option when the amount of bank reserves flooding the system is beyond what they are capable of reeling in.

Hi stone,

Could you comment directly on the Fed's tri-party reverse repo program?  That's the one they're touting expressly as the way to reduce excess reserves.

http://www.bloomberg.com/news/2012-02-2 ... rties.html

Thanks,
KevD

[AgAuMoney]

AgAuMoney

   Increasing reserve requirements DOES reduce the creation of money.  The inverse (reducing) has been demonstrated over and over again.  And if less money is being created, liquidity will be reduced.

I totally agree that increasing reserve requirements does reduce the private bank credit creation of money offset by private debt BUT it does not get rid of bank reserves. Unless I'm in a tremendous muddle with all of this, that is an important distinction. Unless bank reserves themselves get reeled in (putting the toopaste back in the tube) "tight money" won't come about.

"tight money" means money is not available.  The only money available is excess reserves.  Increase reserve requirements and eliminate the excess reserves.  Thus money is not available.  Thus tight money.

As far as the Fed making overdrafts to itself to pay interest on bank reserves- is that really plausible?

Not at all plausible.  You make it too complicated.  The fed has no account to overdraft.  They simply tally the accounts of others, including the federal gov't.  They can credit anything they want to any account at any time.  There would be no overdraft.  For example, when they buy treasuries they simply add $ to the federal account and receive treasuries.  The $ they used to purchase the treasuries did not come out of some mythical fed account, they were sprang into existence the moment they are added to the federal account.

But the interest they pay is a red herring.  The fed only has to pay incentive interest on excess reserves if they want to create excess reserves.  Interest is how they reduce liquidity and incent the banks to accumulate excess reserves.  If, however, they were to raise reserve requirements, they could also reduce interest paid because they would no longer need that incentive.

I suspect the Fed is trying to get to a point when banks will lend to each other again, but until then keeping excess reserves parked at the Fed and keeping a low Fed funds rate is a functional short-term equivalent, albeit accumulating risk for the long term.

[stone]

KevD, thanks for the link. I think you have made me see the light. That reverse repo program is the first example I've been shown and (I hope) understood of how the Fed could create appreciably above inflation short term interest rates at this point within its rules. Am I right in understanding that it enables them to use whatever assets they hold (even long term treasuries) to draw in bank reserves? So let's imagine the Fed has ended up with $1T of LTT and is tasked with reeling back in the $1T of bank reserves that they paid in the first place to buy those LTT. The Fed bought them at an elevated price giving a 3% yield. The Fed wants to create a 6% short term treasury rate. So they do a reverse repo where they promise that the  primary dealer who takes the LTT from them at a $100 price will be able to exchange them back in one years time at a $103 price. So the primary dealer in effect achieves a one year treasury rate of 3%+3%=6%. That all fits in with the Fed rules because the Fed is always allowed to pay too much for stuff. This reverse repo is a way for the Fed to mop up bank reserves by paying even more than they initially did for the assets that the Fed currently holds. I guess the recent reverse repo was on such a tiny scale that it was never intended to actually tighten monetary policy. It was just a proof of principle to display the Fed's powers perhaps to keep a lid on the gold market etc.

AgAuMoney, I thought that the Fed always had to exchange something for any money they created. They have credit creation powers rather than pure base money printing powers. So the Fed could pay $1T for some dubious junk debt or make a $1T loan but couldn't ever just give $1 with no debt attached

[KevD]

KevD, thanks for the link. I think you have made me see the light. That reverse repo program is the first example I've been shown and (I hope) understood of how the Fed could create appreciably above inflation short term interest rates at this point within its rules. Am I right in understanding that it enables them to use whatever assets they hold (even long term treasuries) to draw in bank reserves?

Hi stone,

I'm not sure....I'm a complete novice here.  So far, they've used treasuries, agency, and mortgage-backed securities, but I haven't seen anything about duration.  The tri-party RR "operational readiness test program" was announced well before Operation Twist started and there weren't as many long term Treasuries on their balance sheet back then.  

I guess the recent reverse repo was on such a tiny scale that it was never intended to actually tighten monetary policy. It was just a proof of principle to display the Fed's powers perhaps to keep a lid on the gold market etc.

Yes, just part of their operational readiness test program.  I'm sure it was just coincidental that gold tanked that day.  ::)

KevD

[stone]

http://www.zerohedge.com/article/revers ... requiremen

That article says that the reverse repo idea runs into trouble because bank rules mean that the primary dealers can only take so much Fed assets onto their books before exceeding their Tier1 capital ratio requirement for assets to capital. I guess that is why the Fed has been looking to expand the type of counterparties able to do reverse repos bringing in Fanny and Freddy etc. I guess the bottom line is that as it stands now at this moment, the Fed perhaps could not immediately induce tight money but they are working hard to find a route or at least to look busy about finding a hypothetical route.

[KevD]

http://www.zerohedge.com/article/revers ... requiremen

That article says that the reverse repo idea runs into trouble because bank rules mean that the primary dealers can only take so much Fed assets onto their books before exceeding their Tier1 capital ratio requirement for assets to capital. I guess that is why the Fed has been looking to expand the type of counterparties able to do reverse repos bringing in Fanny and Freddy etc. I guess the bottom line is that as it stands now at this moment, the Fed perhaps could not immediately induce tight money but they are working hard to find a route or at least to look busy about finding a hypothetical route.

Hi stone,

That is an old article.  The Fed has already expanded to Fanny & Freddy and many others.  You can find all the related developments at the following Fed web pages:

http://www.newyorkfed.org/markets/expan ... rties.html

http://www.newyorkfed.org/markets/rrp_a ... ments.html

[stone]

Like the earlier operational readiness exercises, this work is a matter of prudent advance planning by the Federal Reserve. The operations have been designed to have no material impact on the availability of reserves or on market rates. Specifically, the aggregate amount of outstanding reverse repo transactions will be very small relative to the level of excess reserves, and the transactions will be conducted at current market rates. These operations do not represent a change in the stance of monetary policy, and no inference should be drawn about the timing of any change in the stance of monetary policy in the future.

KevD, I suppose it all boils down to whether that expanded collection of counterparties could and would enable a reverse repo that had a, "material impact on the availability of reserves or on market rates".

I suppose the great thing about the PP is that if there is a currency slide and a rate hike can not be achieved to stem it, then gold will (hopefully) save the day and if they do achieve it, then cash will (hopefully) save the day. Perhaps a short term treasury rate hike brought via this reverse repo method would also keep LTT prices high too . My guess is that they won't ever try the short term rate hike for real and the USD and GBP will slowly slip against gold for a very long time but that isn't a guess I lay any store in at all.

[KevD]

KevD, I suppose it all boils down to whether that expanded collection of counterparties could and would enable a reverse repo that had a, "material impact on the availability of reserves or on market rates".

Interesting that reverse repos are being considered right now as part of a sterilized QE program:

http://online.wsj.com/article/SB1000142 ... 82234.html

If you don't have a WSJ subscription, ZH is reporting on it:

http://www.zerohedge.com/news/stocks-pr ... rilized-qe

KevD

[AgAuMoney]

AgAuMoney, I thought that the Fed always had to exchange something for any money they created. They have credit creation powers rather than pure base money printing powers. So the Fed could pay $1T for some dubious junk debt or make a $1T loan but couldn't ever just give $1 with no debt attached

Nope.  Or if you can find a law that says that, you might have a legal case that by violating that law they have reduced the value of your dollar.  Or maybe not.

The Fed has three primary knobs to adjust the money supply:

• reserve requirements
• security purchases/sales
• loans at the discount rate (unsecured, otherwise it is a security purchase/sale)

Until just a few years ago it was assumed that "security purchases" were composed exclusively of open market operations with treasuries.  Now they buy (and maybe someday sell) all kinds of securities on the open market, and under the table, and directly from the treasury.

And while it used to be that they only loaned money to banks in the federal reserve system, now they loan money to banks anywhere of any kind -- commercial banks, investment banks, central banks ...

You can read a whitewashed version: http://dallasfed.org/educate/everyday/ev4.html but if you have been keeping up with current events the past 4 years you will notice many fed actions that page does not really describe.

It was the intention in 1911-1913 to create a central bank with a dual-purpose of full employment and a stable dollar, and give it carte blanche with currency and the banking system to accomplish those purposes.  So that's what they did.  The only real limit in their domain is the 6% limit on profits.  But they get to spend what they want and they aren't ever independently audited, so what kind of limit is that?

Since 1913 the dollar has lost at least 95% of its purchasing power, and we've had the worst periods of economic malaise the country has ever known.

Do you know about FRED?  Can't remember if that's come up in this thread or not.  Found at the St Louis Fed. where they gather all sorts of economic data.  http://research.stlouisfed.org/fred2/ See especially the Adjusted Monetary Base (AMBNS):  http://goo.gl/pgaEW  (sorry for the URL shortener, I hate the things myself but the URL is so ugly that is really the only way to post a direct link to the current graph.)  Another favorite chart is the excess reserves (EXCRESNS).

For both of those charts, they look really impressive with the current spike at the end.  If you end the chart in 2007 or early 2008 you can actually see "normal" operation of the Fed before the huge spike dwarfed anything done previously.

[stone]

AgAuMoney, I thought that the Fed always had to exchange something for any money they created. They have credit creation powers rather than pure base money printing powers. So the Fed could pay $1T for some dubious junk debt or make a $1T loan but couldn't ever just give $1 with no debt attached

Nope.  Or if you can find a law that says that, you might have a legal case that by violating that law they have reduced the value of your dollar.  Or maybe not.

The Fed has three primary knobs to adjust the money supply:

• reserve requirements
• security purchases/sales
• loans at the discount rate (unsecured, otherwise it is a security purchase/sale)

I think we are on the whole sort of agreeing. To my mind neither "security purchases/sales" nor "loans at the discount rate" count as giving money with no debt attached. The Fed can only be benevolent by lending money or paying too much for stuff not by pure something for nothing give aways of say interest on reserves over and above that fundable from the Fed's income stream from its own asset holdings etc. They are not strictly speaking giving something for nothing. I'm not saying that it is anything more than a charade but nevertheless it is a hoop that they restrict themselves to jumping through.
The reverse repo arrangement looks like a way that they might be able to hike short term rates but it will entail depressing long term rates in order to do it. 
Like I said, increased reserve requirements reduce credit creation by private banks BUT does not increase treasury rates. It would increase the interest banks charge for loans but the difference between that and treasury rates can simply widen.

[AgAuMoney]

AgAuMoney, I thought that the Fed always had to exchange something for any money they created. They have credit creation powers rather than pure base money printing powers. So the Fed could pay $1T for some dubious junk debt or make a $1T loan but couldn't ever just give $1 with no debt attached

Nope.  Or if you can find a law that says that, you might have a legal case that by violating that law they have reduced the value of your dollar.  Or maybe not.

The Fed has three primary knobs to adjust the money supply:

• reserve requirements
• security purchases/sales
• loans at the discount rate (unsecured, otherwise it is a security purchase/sale)

I think we are on the whole sort of agreeing. To my mind neither "security purchases/sales" nor "loans at the discount rate" count as giving money with no debt attached. The Fed can only be benevolent by lending money or paying too much for stuff not by pure something for nothing give aways of say interest on reserves over and above that fundable from the Fed's income stream from its own asset holdings etc. They are not strictly speaking giving something for nothing. I'm not saying that it is anything more than a charade but nevertheless it is a hoop that they restrict themselves to jumping through.
The reverse repo arrangement looks like a way that they might be able to hike short term rates but it will entail depressing long term rates in order to do it. 
Like I said, increased reserve requirements reduce credit creation by private banks BUT does not increase treasury rates. It would increase the interest banks charge for loans but the difference between that and treasury rates can simply widen.

My key point is that the Fed creates money when it loans it (lately the discount rate is 0% to 0.25%) or when it buys securities.  It does not loan money "from its own asset holdings" nor does it use its "income stream" to make loans or purchase assets.  The Fed creates money from nothing and distributes it whereever and however it wants.  I'm sure you've seen the "helicopter Ben" references.  Those originated because of Bernanke's paper described how creating and distributing money, even up to throwing it from a helicopter, is within the power of the Fed.

I don't know where your treasury rates statement came from or how it is supposed to apply to anything I've said.  Originally you were talking about reducing liquidity, otherwise known as tightening money, and increasing the reserve requirement is one way to do that.

[moda0306]

AgAu,

I'd agree that the fed's discount window is a form of money creation (even if the bank has to pay it back), but when the fed simply trades reserves for treasury bonds this is not money creation.

The bonds are fiat, government-issued purchasing power... cash is fiat, government-issued purchasing power.  You haven't changed anyone's balance sheet to any great degree.

QE can adjust how people choose to invest, but the banking system tends to kind of do what it does based on demand for loans, and the deficit spending by the federal government is what issues base money into the economy that services those loans.

For a bit of perspective on where we are now... since QE is a quasi-non-event in terms of money-supply, and the discount window is barely used in times of credit contraction (I am pretty sure that it's barely touched even in the past, much less nowadays, in terms of % of total money supply), all we're left with is credit expansion and deficit spending to increase the money supply.

We know credit money is contracting, and that deficits are expanding at over $1Trillion per year.  However, we're "exporting" over $500 billion of our currency to China and other foreigners via our trade deficit, so the actual amount of net money we're adding to our domestic economy through the Government Sector minus the Foreign Sector is probably around $700 Billion per year.

What's credit contraction?  I'm not sure and can't find a solid number, but it would surely drop that $700 Billion to a lower number, would it not?

Let me know where your numbers/analysis disagree with mine.

[stone]

My key point is that the Fed creates money when it loans it (lately the discount rate is 0% to 0.25%) or when it buys securities.  It does not loan money "from its own asset holdings" nor does it use its "income stream" to make loans or purchase assets.  The Fed creates money from nothing and distributes it whereever and however it wants.  I'm sure you've seen the "helicopter Ben" references.  Those originated because of Bernanke's paper described how creating and distributing money, even up to throwing it from a helicopter, is within the power of the Fed.
I don't know where your treasury rates statement came from or how it is supposed to apply to anything I've said.  Originally you were talking about reducing liquidity, otherwise known as tightening money, and increasing the reserve requirement is one way to do that.

I totally agree that the money for Fed loans and asset purchases is created out of thin air. That is also true for everyday bank loans too. If I get a bank loan to buy a car, that too is money created out of thin air. In both cases there is also a debt or asset attached. When the Fed makes a loan, the company that receives the loan owes the Fed money. They have not permanently received something for nothing, they are in debt to the Fed.
My point about reserve requirements and treasury rates was because LTT prices and gold prices are what are relevant to the PP. When people say that the PP is a perilous strategy because "rates will go up", what is relevant are treasury rates. If treasury rates are 0.5% but LIBOR is 5% and junk bonds are 15%, then that is fine for the LTT and gold prices just as much as when treasury rates are 0.5%, LIBOR is 1% and junk bonds are 5%. A Volcker style smackdown of gold and LTT would require the RISK FREE (ie treasury) rates to go appreciably above inflation.

[stone]

Clive, when you say that LTT act as "short gold", I wonder what the consequence would be if this reverse repo Fed action does manage to keep LTT yields steadily inching down