Thursday, September 17, 2015

Post FOMC charts and comments

Going into the FOMC meeting, the market was priced to a 30% chance that the Fed would begin "liftoff." In the aftermath of the Fed's announcement that it would postpone liftoff, short-term interest rates adjusted downwards by about 7-8 bps, which is roughly one-third of a 25 bps "tightening." Translation: the market moved rationally, pricing out what had previously been priced in. Implied volatility and the equity markets jerked around, however, but ended relatively unchanged. So on balance the FOMC announcement was a non-event. It resulted in a very modest reduction in interest rates, but did not increase or decrease uncertainty, leaving the fundamentals reasonably healthy: 2-year swap spreads inched higher to 12.7 bps, but remain very low; gold inched higher but remains in a multi-year downtrend; the dollar inched lower but remains relatively flat year-to-date, and close to its long-term average valuation relative to other currencies; and credit spreads were relatively unchanged while remaining somewhat elevated.

With implied equity volatility relatively unchanged but yields lower, the Vix/10-yr ratio inched higher, but remains significantly below its peak of a few weeks ago. The market is still nervous and still concerned that the U.S. economy is going to prove sluggish, but the outlook is no longer as dire as everyone thought. Equity prices have predictably responded to a reduction in fear and uncertainty by increasing.

Recent changes to gold and 5-yr TIPS prices have been minor, when seen from an historical perspective such as the chart above affords. Both remain in a downtrend, a sign that the demand fro safe assets is declining and that therefore the market is becoming less concerned about drastic outcomes. Fear and uncertainty are gradually being displaced by a slow return of confidence.

The first of the charts above compares the nominal yield on 5-yr Treasuries to the real yield on 5-yr TIPS, and the difference between the two which is the market's expectation for inflation over the next 5 years (1.3%). Not surprisingly, near-term inflation expectations have declined over the past year. As the second chart shows, lower inflation expectations are almost entirely the result of lower energy prices. This also is rational, and does not reflect negatively or positively on the Fed's policy stance.

 The chart above shows inflation expectations for years 5 through 10—currently about 1.9%.

The chart above shows inflation expectations for the entirety of the next 10 years—currently about 1.6%. In other words, the market sees inflation being relatively low for years 1-5 (1.3%), but then picking up to 1.9% for years 5-10, and averaging 1.6% for years 1-10. That's not significantly less than the annualized increase in the CPI over the previous 10 years of 1.95%. As John Cochrane notes,

The outcomes we desire from monetary policy are about as good as one could hope. Inflation is low and steady. Interest rates are lower than Americans have seen in generations. Unemployment, at 5.1%, has recovered to near normal. And banks and businesses sitting on huge piles of cash don’t go bust, a boon to financial stability.

Although far from ideal, things could be a lot worse. There is no evidence to date that the Fed has made any serious mistakes. For newer readers, I note once again that—contrary to popular belief—the Fed has not been printing money, and instead has been merely accommodating a very strong demand for money by swapping bank reserves for notes and bonds.

I continue to worry that when the time comes to raise rates, the Fed will move too slowly, and this will result in an unpleasant increase in inflation and eventually much higher interest rates. But for now this is only something to keep a sharp eye out for. I don't yet see hard evidence of a decline in the demand for money which would precipitate a meaningful rise in inflation, but I have not let down my guard on this. That's the important point: it's all about the demand for money. If the demand for money falls and the Fed fails to counteract that by reducing bank reserves and raising the rate it pays on excess reserves, then inflation and inflation expectations should rise.


Benjamin Cole said...

Scott Grannis: in your estimation, is the Bank of Japan printing money or not?

Scott Grannis said...

I think the evidence shows that the BoJ was way too tight for decades, since the yen appreciated relentlessly against almost all other currencies. They finally corrected that mistake starting about two years ago. I'm not sure that means they started "printing money," or whether they simply stopped making money scarce. Whatever the case, it looks like they are now supplying all the money the economy demands. The yen is trading very close to its PPP level, by my calculations, and the stock market has risen significantly, even in dollar terms. To me, that suggests that monetary conditions in Japan today are just about right: neither too easy nor too tight.

Richard H. said...

Wait a second, "swapping bank reserves" implies the money was already there, sitting around in the Fed's vault. But it wasn't. It was newly created (printed).

Benjamin Cole said...

Scott: Yes---and the Bank of Japan is conducting about $80 billion a month in QE, if my memory serves and the exchange rates are what I think. This leads me to suspect QE should be considered conventional policy.

Richard H. said...

Also, please, what do you mean by "decline in the demand for money which would precipitate a meaningful rise in inflation"?
Things are getting just too weird: isn't there always an unlimited demand for money, at least at zero or very low interest rates?

Scott Grannis said...

The Fed can't print money, but it can create bank reserves anytime it buys notes or bonds. For the Fed, the purchase of assets is equivalent to a swap, or an exchange. The Fed credits banks with reserves, and in exchange receives notes and bonds. From the perspective of banks, they sell notes and bonds to the Fed and receive bank reserves, which are functionally equivalent to T-bills since they are risk-free and short-term interest-bearing assets. The transaction is also equivalent to the Fed borrowing money from the banks to buy notes and bonds; the interest rate the Fed pays on the loan is equal to the interest it chooses to pay on excess reserves.

Scott Grannis said...

As Milton Friedman taught us, inflation happens when the supply of money exceeds the demand for it. To date, the Fed has created trillions of dollars of "money" (i.e., bank reserves, which are functionally equivalent to T-bills and which are the ultimate form of money, since they have no risk, are liquid, and pay interest. Banks have to date been happy to hold on to those reserves and have used only a small fraction of them to collateralize new lending. If banks should ever decide that those reserves are not paying sufficient interest (if they can make more money on a risk-adjusted basis by lending to the private sector), they will start ramping up their lending and thereby increase the supply of money in the economy. This is the mechanism by which inflation could begin to rise, because there would effectively be a surplus of money relative to the demand for it.

Benjamin Cole said...

I checked the figures. The Bank of Japan does about $55 billion a month in QA, in an economy about half the size of the United States.

Scott Grannis said...

It's important to remember, however, that QE is not equivalent to "printing money." Based on our experience so far with QE, it is better thought of as satisfying the demand for money equivalents.

Richard H. said...

There are so many things that don't make sense (that seem twisted to me). First, what is the meaning then of the fed's balance sheet? By your logic, it doesn't matter how large it gets; it's all just a transfer. Second, wouldn't banks always be willing to sell more bonds and notes at very, very high prices to the Fed in favor of an almost equivalent interest rate - riskless short term T-bill (bank reserves)? Third, you say "If banks should ever decide ... they can make more money ... by lending to the private sector, they will start ramping up their lending and thereby increase the supply of money in the economy. This is the mechanism by which inflation could begin to rise, because there would effectively be a surplus of money relative to the demand for it." This does not make sense. The reason, as you say, they are ramping up lending, is because there is a demand for it. Then you say that inflation will happen because there is not (enough) demand. Why would they ramp up more than there is demand for? If there is some reason why aren't they doing it now? And really, who makes it possible for the banks to ramp up lending - that causes inflation - excess bank reserves from the Fed (if I am understanding this).

Finally, why do banks have to be involved entirely in the Fed's balance sheet? The Fed has been buying bonds basically directly from the government; the government takes the money, yes, and deposits it at a bank I guess, but then spends it. How else can this possibly be looked at?

I appreciate your earlier response and would very much like to have your thoughts.


WilliamSSmith said...


This is unrelated to the topic at hand but I wanted to make a point about something you said a couple of years ago about the Aussie buck. At that time I suggested that the A$ was overvalued at over 1:1 and you kindly analyzed the situation based on a PPP analysis and suggested that the A$/US$ would eventually go to .67. I am happy to report that I am here (in Oz) and my U.S. dollars again feel like real money -- with the A$ at about .7. Well done.
Bill Smith

Hans said...

Reserves are in effect walk-a-round money or wallet money, in
case there is a need for "fast" cash.

I have asked this question before and never get an answer.

Why would a banker trade notes and bonds for lower yielding

The EXPLOSION of M1, since you guessed it, 2008.

Mike said...

Hi Scott,

I thought they were keeping all the interest - like 600 Billion so far and giving it to the Fed to pay down debt. I also thought the Fed said they would not unwind the transaction and are keeping all the runoff as the debt matures / mortgages prepay.

Isn't that the equivalent of printing a lot of money - short and long term?

Thanks - Mike

Ps - Are you concerned all the Federal debt is financed 4yrs. and under at .25bps and is 6% of Gov. spending... how do you think that unwinds?

Johnny Bee Dawg said...


It's been answered over and over.
Risk aversion.

Have you ever seen a chart of M2?

Hans said...

Mr Dawg, there is risk aversion in federal notes and bonds..LOL

And yes I have seen a chart of M2, which merely adds to the totals of
seen charts of M1.

I am still waiting for answer and more than likely will forever because
there is no sound reply, other than the FRB telling bankers do it or else.

Hans said...

Regarding the effects to the money supply from FRBQEs..

"February 16, 2015
One of the most persistent beliefs in the world of economics today is that the Fed’s QE (Quantitative Easing) adds to bank reserves but does not directly boost the US money supply. The popularity of this belief is remarkable considering that anyone who bothers to do a few simple monetary calculations will quickly see that it is completely wrong. The fact of the matter is that every dollar of QE adds one dollar to bank reserves AND one dollar to the economy-wide money supply.

Before I briefly explain the process by which the Fed’s QE injects money directly into the economy, I’ll show the simple calculations that anyone commenting on monetary matters should do. The calculations are based on the fact that new US dollars can only be legally created by the commercial banking system and the Fed.

The commercial banks create money when they make loans or monetise assets. More generally, commercial banks create money via an increase in credit. The increase in total Bank Credit over a period is therefore a rough, but reasonable, estimate of the MAXIMUM amount of new money that could have been created by the commercial banking system over the period. Note that changes in Bank Credit are recorded in the Fed’s H.8 Release.

At the end of August-2008, which was just prior to the start of the Fed’s first QE program, total Bank Credit was around $9T (9 trillion dollars). At the end of January this year it was around $11T. This means that the commercial banks have collectively created a maximum of 2 trillion new dollars since August-2008. They might have created significantly less than 2 trillion new dollars, but they have not created significantly more than that.

Let’s now consider what happened to US True Money Supply (TMS) over the same period, noting first that TMS is the sum of physical currency in circulation, demand deposits at private depository institutions and savings deposits at private depository institutions. TMS only counts money within the economy. It does not count bank reserves.

From the end of August-2008 through to the end of January-2015, TMS increased by $5.1T. Since we know that commercial banks created a maximum of $2T over this period, we know that at least $3.1T came from somewhere other than the commercial banking system. And since we also know that new US dollars can only be created by the commercial banks and the Fed, we therefore know that the Fed’s QE must have directly created a minimum of 3.1 trillion new dollars.

I’ll now move along to the process by which the Fed’s QE boosts the money supply.

The first point that must be understood is that the Fed conducts its asset purchases and sales via Primary Dealers (PDs). In many cases the PDs are banks, but in such cases the PD part of the business is separate. Of particular relevance, the PD part of one bank will maintain demand deposit accounts at other banks and these demand accounts receive the payments when the Fed buys assets from the PD.

Hans said...


Next, for the sake of explanation let’s assume that PDA (Primary Dealer A) is a subsidiary of Bank A and maintains a demand deposit at Bank B. When PDA sells assets to the Fed, the Fed deposits payment in the form of newly-created dollars into PDA’s demand account at Bank B. Since customer deposits are liabilities of banks, if the process ended with the Fed depositing new dollars in PDA’s account at Bank B it would increase Bank B’s liabilities by the amount of the deposit. To make the process balance-sheet-neutral for Bank B and the banking system as a whole, the same amount that was deposited in PDA’s demand account at Bank B is added to Bank B’s reserves at the Fed. In effect, the Fed adds dollars to demand accounts within the economy that are covered by reserves at the Fed.

One dollar of QE therefore involves one dollar being added to a demand deposit within the economy (part of the money supply) and one dollar being added to a reserve account at the Fed.

Let’s now take a look at how the mechanics of the QE process as outlined above explain the change in the money supply since the beginning of the Fed’s QE back in 2008.

I mentioned above that if we only consider the amount of money created by the commercial banks then we find that at least $3.1T is unaccounted for. If my analysis is correct then the Fed’s QE must have directly added a minimum of $3.1T to the money supply.

A very rough approximation of the amount of new money added by the Fed over a period is the change in Reserve Bank Credit, which can be determined by referring to the Fed’s H.4.1 Release. The increase in Reserve Bank Credit from August-2008 until January-2015 was $3.6T, which is in the right ballpark. However, a more accurate calculation of the amount of new money created by the Fed can be done using the knowledge that a) each new dollar added to the economy by the Fed will be associated with one dollar of additional reserves, and b) reserves at the Fed will remain at the Fed unless they are removed by the Fed or they are converted into physical notes/coins (in response to increased demand by the public for physical currency). The amount of money created by the Fed since August-2008 should therefore be equal to the net increase in Non-Borrowed Reserves at the Fed plus the increase in Physical Currency in Circulation over the same period.

The figure comes to $3.4T, which is roughly what it needs to be to explain the increase in True Money Supply.

Hans said...

"The figure comes to $3.4T, which is roughly what it needs to be to explain the increase in True Money Supply.

A separate question is: Why hasn’t the large Fed-promoted increase in the US money supply led to a substantial increase in the ‘general price level’?

This is a question for another time as this post is already too long, but suffice to say right now that:

1) There has been a significant increase in the ‘general price level’ as a result of the monetary inflation, just not as significant as would normally be the case.

2) The general price level’s smaller-than-normal response to the money-supply increase of the past several years is probably related to the Fed’s abnormally-large role in the money-creation process. During more normal (pre-2008) times, almost all new money is created by the commercial banks. Consequently, the first receivers of the new money tend to be within the ‘general public’ (home buyers/sellers, private businesses, etc.). However, during the period since August-2008 about two-thirds of all new money has been directly created by the Fed. This means that the first receivers of most of the new money have been bond speculators, and that the second, third, fourth and fifth receivers of the new money have probably been bond speculators or stock speculators.

In conclusion, when I say that the Fed’s QE directly boosts the money supply I’m not stating an opinion or giving my interpretation of how the monetary system works. I’m stating a fact."

Scott Grannis said...

Mike: re Fed's interest earnings. The Fed earns interest (~1.5-2%)on the trillions of notes and bonds it owns, and pays 0.25% on the bank reserves it created to buy those notes and bonds. The difference is the Fed's "profit," which amount, after paying its own expenses, the Fed remits to Treasury. Last year that amounted to about $100 billion. This year will probably be similar.

Scott Grannis said...

Hans: The article you quote is incorrect. The Fed's purchases of notes and bonds are financed by the issuance of bank reserves. The Fed can only create bank reserves; the Fed can't create money unilaterally or directly. The Fed issues new currency on demand from the banking system, but banks must surrender $1 of reserves for each $1 of currency. QE does not boost the M1 or M2 money supply directly. QE does boost the monetary base, of course, since that the base is defined as bank reserves plus currency outstanding.

Benjamin Cole said...

My toupe and cane nod in agreement with you!

Thinking Hard said...

All of this talk about raising rates and whether it is the right time or not. Everyone is focused on the wrong question. Let’s talk about what is “normal”. I hear a lot of media pundits talking about the normalization of interest rate policy. That is a red herring if I have ever seen or heard one.

Federal Reserve balance sheet (All Federal Reserve Banks – Total Assets, Eliminations from Consolidation): September 3, 2008 = $905B vs. September 2, 2015 = $4.475T

Excess Reserves of Depository Institutions: August 2008 = $1.875B vs. August 2015 = $2.512T

Interest Rate Paid on Balances Maintained that Excess the top of the penalty free band (IOER): August 2008 = 0% vs. August 2015 = 0.25%

U.S. Treasury securities held by the Federal Reserve: September 3, 2008 : $479B vs. September 2, 2015: $2.46T

Approximate value of UST maturing in: 2016 - $200B; 2017 - $200B; 2018 - $400B

Federal Reserve normalization procedures during the rate hike cycle include raising IOER and ON RRP so the FFR falls in between. The FFR will likely trend towards the bottom of this range. This is why when I see major economists predicting a firm 0.25% increase that is just wrong and shows a lack of understanding of the process. The FFR is not set at a given rate but instead is given a target range.

Previously, the FOMC would undertake an interest rate increase through buying/selling on the open market. This time, the Federal Reserve plans to rollover maturing UST debt, attempting to maintain a more static balance sheet.

I stated this months ago, and still believe it is pertinent now. The Federal Reserve is hesitant to come off the bottom due negative externalities found in international markets and the unknown consequences of beginning the utilization of new tools.
Let’s focus on the elephant in the room, or potentially the bear in the corner, and ask the more pertinent question: What is “NORMAL”?

Interesting moves in the dot plot seen here (
One board member sees the FFR needing to be NEGATIVE in 2015 and 2016. Why? Implications of a negative FFR? First time I have seen a board member show a need for negative FFR rates.

Hans said...

Mr Grannis, I will reread your post and the article I copied and pasted.

Thank you for your thoughts.

Ben Jamin, you are the best!!

Charles said...

There is no reason to worry about the business cycle. The economy remains deeply depressed. Real growth is low, employment is low. Investment is low. Housing starts are low. This is stagnation. And it is a predictable consequence of Obama's micro-economic policies.

As to inflation, price indexes are rising slowly but I see little evidence that anything the Fed has done or failed to do has had any effect on inflation. This I see as evidence that the Fed today has no control over any meaningful measure of money.

The Fed should stop trying to manipulate interest rates entirely.