Breaking News! QE II is not ending!

The "End of QE II" story has turned into the biggest source of market misperception in decades.  I watch and read financial news constantly.  Each day goes by with no one correcting the many factual errors on this subject.

The QE II matter is probably the most important market issue, so getting it right should have a big payoff.  The story is too big for one of my regular evening articles, so I will continue with my planned series.  Tonight's subject is how QE II really works and why it is not ending.

I realize that there are many other questions -- the effect of market perceptions, the apparent, but deceptive correlations, and even the effect of misperceptions.  I will take up these questions in due time.  For the moment, you can check out a summary of my past QE II articles.

A Simple Illustration

Let me start with a simple and more traditional illustration of Fed policy -- cutting interest rates.  As the Fed reduced rates, we would say that it was a cycle of easing.  If the Fed kept rates at a very low level, we would describe it as a continued easy money policy, not the end of easy money.

The quantitative easing programs took place when the Fed had already exhausted the traditional rate cutting at the short end of the yield curve.  The Fed will complete the planned purchase of $600 billion in Treasuries by the end of June.  The purchases are the equivalent of interest rate easing.

Stopping Fed purchases does not end the period of easy money.  That will happen only when the Fed reduces its balance sheet, something that is not currently planned.

To emphasize -- the market believes that the effect of the Fed policy comes only from fresh purchases of Treasuries.  This incorrect conclusion has been widely disseminated by various sources, including those who should know better.

The reality is that the easy money policy continues as long as the Fed balance sheet is not reduced.

Explanation

To understand this very important distinction, you need to know how a fractional reserve banking system works.

How it does NOT work

What is definitely not happening is the thing you hear and read every day.  People think that the dollars from the Fed buying of Treasuries somehow finds its way into stocks and commodities and has fueled a speculative bubble.  This is a know-nothing attitude, which you will easily understand if you think about it carefully and objectively.

The Fed buys bonds from a dealer.  From the dealer's perspective, the transaction in no way adds to the ability to make other purchases.  If a dealer bank wanted to buy grain or stocks or oil futures, this can be done with very little margin and using bonds as collateral.  The Fed transaction has no effect.

How it does work

The Fed purchase of Treasuries from dealers increases the monetary base.  The money creation effect -- and this is the "printing money" you hear so much about -- is the reciprocal of the reserve requirement, currently ten percent in the US.  This means that $600 billion in QE II  buying has created the potential for $6 trillion in new money.  This potential has not been realized.

The Wikipedia articles are fairly accurate but slender introductions that provide a starting point (here and here).  The best explanation of QE II is Mark Thoma's description.  It is worth your time to read the full article, but here is the key conclusion:

Since it is predominantly long-term rates that determine business investment, the purchase of new homes, and the purchase of consumer durables such as cars and refrigerators, this was of concern within the Fed. But the reason for this was never fully understood, and the crisis diverted attention away from this issue. However, one way the Fed can potentially overcome this problem is to buy and sell longer term Treasury bonds, i.e. those that exist on the long end of the yield curve, to bring long term rates up or down as desired.

This is, essentially, all that QEII is. It is conventional monetary policy that operates at the long end of the yield curve through the buying and selling of long-term financial assets rather than through the more traditional buying and selling of short-term assets.

The Effect of QE II

It is pretty much impossible to know what the effect of QE II has been.  The key question is what interest rates would have been in the absence of this policy.  Most economists estimate that the rates would have been about ten basis points higher in the middle of the curve, but no one can prove the case.  In fact, rates moved higher after the policy was implemented, a deceptive but seductive estimate of the effect.

Other effects came from psychology.  More on that in future articles.

The actual effect comes from increased bank lending.  From the Fed's perspective, the policy did not sufficiently stimulate lending. The result was a reduction in the velocity of money, explained effectively (and with great charts) by two of our favorite professors, James and Menzie at Econbrowser.

I made similiar arguments in refuting some of the worst commentary on the subject.

The End of QE II?

One would think that the Fed Chairman would be the best source on whether QE II is ending.  Ben Bernanke covered this question carefully in the Q&A that was part of his recent press conference.  The answer is technical and seemed to escape the notice of most.  Here is the first part of his answer:

CHAIRMAN BERNANKE. As I—as I’ve noted and as you’re all aware, we are—we
are going to complete the program at the end of the second quarter, $600 billion. We are going
to do that pretty much without tapering. We’re just going to let the purchases end. Our view is
that—based on past experience and based on our analysis—is that the end of the program is
unlikely to have significant effects on financial markets or on the economy, the reason being
that, first—just a simple point that we hoped we have telegraphed today—we hope we have
communicated what we’re planning to do. And the markets have well anticipated this step. And you would expect that policy steps which are well anticipated by the market would have
relatively small effects because whatever effects you’re going to have would have already been
capitalized in the financial markets.

I note that Bernanke seems to be out of touch with market anticipations!

Let us turn to his real expertise [emphasis added]:

Secondly, we subscribe generally to what we call here the stock view of the effects of securities purchases, which—by which I mean that what matters primarily for interest rates, stock prices, and so on is not the pace of ongoing purchase, but rather the size of the portfolio that the Federal Reserve holds. And so, when we complete the program, as you noted, we are going to continue to reinvest maturing securities, both Treasuries and MBS, and so the amount of securities that we hold will remain approximately constant. Therefore, we shouldn’t expect any major effect of that. Put another way, the amount of ease, monetary policy easing, should essentially remain constant going forward from—from June. At some point, presumably early in our exit process, we will, I suspect, based on conversations we’ve been having around the FOMC table, it’s very likely that an early step would be to stop reinvesting all or part of the securities which are coming—which are maturing. But take note that that step, although a relatively modest step, does constitute a policy tightening because it would be lowering the size of our balance sheet and, therefore, would be expected to essentially tighten financial conditions.

To emphasize, the halt in new buying is not tightening, especially if maturing securitites are reinvested.

Investment Conclusion

I cannot do the full investment conclusion from this single article.  We must consider the perceptions as part of the effect.  Some perceptions were intended by the Fed.  Others are completely mistaken.

This is an important opportunity for investors who take the time to understand the issue.  As I noted in our weekly update, it is a time to pick and choose great stocks at great prices over the coming month.

A Final Note

QE II will end when the purchases are unwound.  There is an important difference between the end of new purchases and the end of QE II.  It is easy to be sloppy in discussing this, and everyone (including me) lapses into the popular perception on occasion when we should be much more careful.

Comments

[Steel Phoenix]

There's something here that I'm missing I think. I'm no expert on the matter, so maybe you guys can help me out.

The article covers things from the standpoint of the Fed, but what about the Treasury? In the short term, if the government wants to pay its bills, won't it either have to sell assets, print money, or sell treasuries? With the Fed out of the picture, won't they have to raise the interest on treasuries in order to attract buyers?

[oldprof]

kharris -- Sorry to get your argument wrong, and thanks for clarifying. I agree with you. We will find it just as difficult to measure the effect of the "end" of QE II as we have in measuring the effects so far.

As to the "quibbling over words," that originally caused you to write -- You have stated your point clearly, and I think we all get it. My contention is that the terminology that nearly everyone is using is misleading, since it focuses on the (unimportant) buying and ignores the (very important) balance sheet.

This is an important substantive point which hardly anyone understands. Happily, most people seem to have grasped the significance of the argument and are untroubled by the dramatic nature of my chosen title.

BTW, James Altucher did me one better tonight on Kudlow. He stated that QE II has not even started!!

Thanks again for joining in.

Jeff

[kharris]

Rich,

Just right on the risk of the Treasury curtailing issuance. Our host seems not to have understood that point when I made it.

The quality swap point you make is also correct. I'm not sure your point about maturity is. While Fed purchases relative to market flow are small, they have had a much larger impact on the stock of outstanding Treasuries and the maturity.

[kharris]

Jeff,

Your argument is based on insistence on your own definition. "The QE II program will...end when the assets are sold." This is the claim you make to argue that QEII isn't ending in June. This is pure question-begging, nothing more than you saying that the "end" is when the assets are sold rather than when purchases stop.

Your argument is based on a claim about how the language should be used. In such an argument, you can reasonably go three ways. You can rely on historical usage, but there is not much established history because "QEII" is a neologism. You can rely on common usage. You have provided the right answer in terms of common usage: "I am sure that your perspective is shared by most." If so, then according to common usage, your view is wrong. The third way to go is to pretend that one is the Queen of Hearts and that words mean what you want them to mean. So far, you seem to be taking the Queen of Hearts approach.

As to this: "I do not think that the end of these purchases will have any effect at all on the size of Treasury auctions. We will not have long to wait to discover which of us is right about that one!"

-- you have misconstrued what I wrote. I did not claim that the end of Fed purchases would change the size of Treasury issuance. I wrote: "By coincidence, the Treasury will very soon have to limit the sale of new Treasury debt..." The word "coincidence" in no way implies a causal relationship. I'm not sure how you concluded that it did. My point, which I hope I made clear, is that the possible near coincidence of the Fed ending its purchases and Treasury curtailing sales will make it difficult to tell how much impact the end of Fed purchases has on market pricing. You would be quite mistaken to claim to be "right about that one" unless, in the case that Treasury does have to curtain issuance, you can show that there is no confusion about the impact of the end of Fed purchases on market prices.

If you want people to take your writing seriously, you need to show that you take their writing seriously as well. I don't think I write so badly that what I wrote could reasonably be construed to mean what you have taken it to mean.

[RichL]

To put a different spin on all this, isn't the govt. simply doing a quality swap; upgrading portfolio by selling Maiden Lane securities, AIG, and other assorted detritus, and buying Treasuries? And the dollar amount of QE2 relative to the daily volume of Treasuries is not very large. The whole thing's a non-issue.

More ironic is that forward Treasury issuance will be rather small if there is no agreement on the debt limit extension. It's hard to have a bear market in a security where the issuer is constrained from issuing!

[Andrew]

It would be more accurate if you would switch that to a log scale. Once you do that, the growth rate is contant, which is what I was saying. The average growth rate since the 1960s is 6.9% and for each decades 1960 - 7, 1970 - 9.5, 1980 - 8, 1990 - 4, 2000 - 6.5. 2010 - 2.7. So the current growth is not even at average.

I feel like I shouldn't be commenting because 95% of message boards seem to be from doom and gloomers. Yes, a broken clock is right twice a day.

[oldprof]

Mike -- Thanks for the kind words. As to your quibble -- Margin rates are not a function of Fed purchases. They are based upon exchange rules concerning volatility and risk.

I do agree that the program kept interest rates slightly lower than we would have experienced in the intermediate part of the yield curve.

The direct impact on asset purchases is too small to measure. They psychological impact? Maybe....another topic.

Thanks for joining in.

Jeff

[oldprof]

kharris -- The QE II program will reach its maximum size during the next month. It will end when the assets are sold. As BB has made clear, there is not even a time frame for ending the program.

My principal point, which most people do not understand, is that the effect comes through the impact on the money supply (which continues) rather than from the asset purchases.

I do not think that the end of these purchases will have any effect at all on the size of Treasury auctions. We will not have long to wait to discover which of us is right about that one!

Thanks for taking the time to write. I am sure that your perspective is shared by most. It certainly has gotten the most attention in the last few weeks.

Jeff

[kharris]

Exactly. Accounts which hold Treasuries will, in some cases, not hold commodities. They will, however, hold other financial securities such as stocks and corporate bonds and foreign debt. It is easy to say that accounts "could" use Treasuries as collateral for commodities, but we care about what they do, more than what the could do.

[kharris]

John,

My error was to make use of standard Treasury market slang for an audience, at least as you represent it, that is unfamiliar with that slang. In standard industry lingo, "coupons" are coupon-bearing securities, as distinct from bills or discount notes ("discos" for those of you who dance). Glad to have been able to clear that up for you, and sorry to rob you of your chuckle.

[Mike]

Great stuff here! I do quibble with the "not" part of the article.

"People think that the dollars from the Fed buying of Treasuries somehow finds its way into stocks and commodities "

Do you not think FED purchases have kept margin rates low? Lower funding costs from dealers has always tempted investors to leverage up and chase yield/risk

[wsm]

kharris has it right. Old Prof's post does not explain how "QE2 is not ending!" If anything, I am now more convinced that it is indeed ending.

As kharris correctly points out: 'Your argument, though, only shows that the Fed will not end "quantitative easiness". It certainly will stop "easing", taking active steps to increase accommodation. The Fed will be "easy"' but not "easing", and the "E" in "QE" is "easing".'

[John Forman]

kharris - Can't help but chuckle at the irony of you arguing about terminology when you've made a blatant error of your own. :-)

The Fed isn't buying Treasury "coupons". Those are the individual periodic interest payments. The Fed is buy Treasury securities - Notes and Bonds.

[Philip George]

Why do you even bother to look at M1 or M2? A monetary aggregate that rises even more sharply during a recession than otherwise (see http://research.stlouisfed.org/fred2/series/M2NS?cid=29) is obviously worthless.

Look at my graph and note how accurately it tracks the period before the Great Recession and the Great Recession itself.

It shows that Corrected Money Supply at the end of February was at a level dangerously close to its peak at the beginning of 2006, indicating that a financial crash is almost certainly imminent.

[Andrew]

The money supply has been rising, but not nearly at the same pace as the monetary base. Since the third quarter of 2009, money supply (M2) is up 5.4% and is up 6.1% (non annualized) since March of 2009. The monetary base (Fed's balance sheet) is up 39% and 52% over the same period, respectivly. This has resulted in the money multiplier dropping to a record low of 3.7 compared to its long term average of 9.4. In other words, the money the Fed is "printing" is not making its way into the economy because banks are not lending at the pace they have historically.

Getting back to the growth in money supply, what we are having now is actually lower than the historical average 1-year growth rate of 6.9%.

Said another way, this is not inflationary unless banks step up their lending to historical norms.

[kharris]

This wonderful insight you offer strikes me more as quibbling over what words are permissible than an actual contribution to understanding. There may be some folks - those who aren't really following Fed actions anyway - who don't know that the Fed intends to stop buying large chunks of Treasury coupons or don't know they do not yet intend to sell large chunks of Treasury coupons. Among those who are watching, those simple facts are understood.

First off, I think you argument is wrong linguistically, and since I also think you are quibbling over words, getting the words wrong means you are utterly wrong. All you have done is told us that you think we should no call this and "end" to quantitative easing, which is what "QE" is generally taken to mean. Your argument, though, only shows that the Fed will not end "quantitative easiness". It certainly will stop "easing", taking active steps to increase accommodation. The Fed will be "easy"' but not "easing", and the "E" in "QE" is "easing".

When it comes to substance instead of language, it's hard to tell whether you have your story right, because your story ignores a good bit of the substance. The Fed's analysis finds that the greatest impact from asset purchases is through the reduction in the stock of Treasuries held in private hands. Whatever model one may use to describe the workings of asset purchases, that is what the Fed's empirical research has uncovered. The Fed will no longer be doing what it has determined to be the most powerful thing it can do once short end rates reach zero. You've left that entirely out of your story.

By coincidence, the Treasury will very soon have to limit the sale of new Treasury debt soon after the Fed stops buying it. Treasury is taking over on the supply side the job that the Fed is leaving off on the demand side. That is going to muddle the result of the Fed leaving off.

[Philip George]

You say:
The money creation effect -- and this is the "printing money" you hear so much about -- is the reciprocal of the reserve requirement, currently ten percent in the US. This means that $600 billion in QE II buying has created the potential for $6 trillion in new money. This potential has not been realized.

How do you know the potential has not been realized? On what parameter are you basing your conclusion?

My calculations show that money supply has been rising constantly since the third quarter of 2009.

The graph can be seen at http://www.philipji.com/item/2011-05-11/how-much-longer-until-a-crash