Wednesday, September 29, 2010

Thoughts on quantitative easing


Ever since the Aug 10th FOMC statement—in which the Fed announced it would be buying longer-term Treasury securities with the proceeds of its maturing or prepaid Agency and MBS holdings—there has been a very interesting and tight correlation between the slope of the Treasury yield curve from 10 to 30 years and the market's own inflation expectations. This is shown in the above chart, with the red line representing the 5-yr, 5-yr forward inflation expectations embedded in TIPS securities, and the blue line representing the slope of the Treasury yield curve from 10 to 30 years.

What stands out is that the slope of the longer end of the yield curve is now a good proxy for the market's inflation expectations. That is at it should be, of course, since the higher inflation expectations, the greater the premium that investors should demand to own 30-yr bonds instead of 10-yr bonds. But it hasn't been that way for some time. And as the next chart shows, the slope of the 2-10 portion of the yield curve has been trending down all year even as inflation expectations have perked up. Plus, the flattening of the 2-10 portion of the curve has occurred under very unusual circumstances. (Typically, the curve flattens when the Fed pushes up short-term rates, and it steepens when the Fed lowers rates. For some time now, the Fed has kept short-term rates steady at very low levels, while longer-term rates have been falling.)


So the behavior of the yield curve is telling us something important, namely that the Fed's purchases of (and intention to continue purchasing) longer-term Treasury notes is having an impact. The Fed is artificially depressing yields out to 10 years, and that's not surprising because that's what they are aiming for. The Fed believes that lower long-term yields will be stimulative for the economy.

Whether the Fed's purchases of bonds will prove to be a stimulus for the economy remains to be seen, of course. Since early August, lower 5-yr and 10-yr Treasury yields have not resulted in any significant decline in mortgage rates, for example, because the spread between Treasuries and mortgage rates has simply widened. This is not unusual at all, it is simply the market saying that it doesn't believe lower Treasury yields are permanent, and/or it doesn't think that buying mortgages at lower yields is likely to prove profitable. And even if the Fed were able to drive mortgage rates to artificially low levels, I think it's questionable at best whether this would prove to be a stimulus for the economy.

Artificially low borrowing costs are part of the reason we're in the mess we're in. Cheap credit, among other things, helped fuel the housing boom, which eventually went bust. Flooding the system with money could help bail out underwater homeowners by pushing up home prices, but only at the cost of another round of reflation (perhaps housing prices, or in some other area of the economy, who knows?). Plus, it's hard to convince people to borrow these days, when so many are still smarting from having borrowed too much some years ago.

But I suppose that if the Fed tried hard enough for long enough, it would soon become apparent to intelligent people that taking out a whopping big mortgage was a good way to become rich. Borrow now at a super-low fixed rate for 30 years, buy a bigger home or some other tangible asset, then sit back and wait for the price level to rise and reduce the cost of repaying your loan. If enough people decide to borrow more, that translates into a reduction in the demand for money, and that has the effect of increasing the amount of money in the system relative to the prices of goods and services. It shouldn't be hard to see how that would in turn result in a higher price level for just about everything. It won't, however, result in any material change in the economy's ability to grow, since growth only occurs when the productivity of labor rises—when we collectively produce more for a given amount of effort.

I think the bond market is already thinking along these lines, and that is why the long end of the yield curve is steepening. The Fed may be able to depress 10-yr yields by promising to keep the funds rate at zero for an extended period of time, but there is no way the Fed can convince investors to buy 30-yr bonds a ridiculously low yields. Savvy investors are figuring this out: quantitative easing is going to push up inflation, so the thing to do is to shun long-term bonds (or borrow at long-term rates), and buy tangible assets or other currencies. Did I mention that gold and other currencies are already rising? And that's why the steepening of the long end of the curve is indeed a good sign that quantitative easing is going to lift inflation.

Along the way to higher inflation—which could take years to show up—this Fed exercise in quantitative easing may have at least one salutary effect, and that will be to vanquish the widespread fears of deflation. Convincing people that holding onto cash yielding zero is a bad idea is one way of boosting the velocity of money, and that is in turn a way of boosting the economy, if only because velocity has been very depressed. People have been hoarding money since the financial crisis erupted, and the hoarding continues to this day. That has depressed growth in the economy, which is another way of saying that fear of the future and risk aversion are not compatible with healthy growth.

I'm not condoning a QE2, however. I am hopeful, in fact, that it will not prove necessary, and I think that will become obvious as more signs of economic growth show up in coming months.

19 comments:

Benjamin Cole said...

Excellent commentary.

If even Scott Grannis says it will be years until the possible inflationary impact of QE is felt (if then), then I say damn the torpedoes and full-steam ahead on the QE2.


There are times when we must suffer to get to a greater good--delayed gratification etc.

Then there are times we suffer, just so we can suffer. We are in such a time now.

Inflation is dead. Banks need their real estate loans to make good (or better) in next few years. Reflate those properties pronto.

Did anyone in Japan 20 years imagine they would go into a 20-year-long low inflation-deflation funk, and see property markets and equty market tank by 75 percent?

Milton Friedman advised Japan to engage in QE until they generated inflation. They did not listen.

The yen has been so strong throughout. Conclusion: Tight money is misery for equity and property owners. Zero inflation is a dangerous utopian pipedream.

BTW, Bernanke says it was not cheap money but huge capital flows into housing, through Fannie, Freddie and the private MBS market, and low underwriting standards, that boomed residential real estate. I agree. Capital is everywhere.

Some of us older fogies probably think of capital as being scarce, and remember the days of "crowding out."

But mature economies, such as Japan, and Europe, and now the USA, seem to pass into a stage in which capital is not scarce, even if interest rates get low. Higher income people can save more, and middle-income people have to save for retirement, regardless of interest rates. Capital markets are globalized.

The Treasury can raise money easily, any good business can raise money easily. See Petrobas and their $70 billion raise.

I am hoping good times are right ahead, if only the Fed would let the monetary bulls run the place for a while.

"Profiles in Timidity"--not how successful policies are made. The Fed is being very timid.

PS Of course, simple monetary demand does not boost productivity. It does boost deamnd and output when there is slack, and we have now have slack, gobs of it.

I trust American businesses will do what they always do, and invest in new plant and equipment when there is good aggregate demand.

I don't blame businesses for not investing when demand is limp, and deflation lurks.

W.E. Heasley said...

Mr. Grannis:

Excellent analysis.

However, viewing QE2 from a political-economy perspective, we might be well advised to remember this compact, simple, and highly informative observation by Thomas Sowell:

- governments for generations have transferred wealth from the people to themselves by simply issuing inflationary amounts of newly created money and spending on what ever the government wants to finance -

randy said...

Fantastic commentary Scott.

Benjamin,

Yes inflation could reflate asset prices and save banks and homeowners... for a while. But we've proven debt driven demand eventually collapses. Business investment made to meet debt driven demand eventually collapses. We need sustainable demand driven by productivity (per capita income) and population growth and exports. I don't know the answer to that but I don't think it's QE2. Getting back on the roller coaster might just make me vomit.

Also, as Yardeni recently wrote:

"My theory is that the Keynesian apparatchik in both Japan and the US welcome economic and financial crises as great opportunities to grab power. So they come to our rescue with massive spending programs financed with lots of borrowed money. The Japanese government built roads and bridges to nowhere that nobody needed. The US government can’t seem to even do that. Instead, the stimulus spending has been focused on keeping unionized public workers employed. The government’s intrusion into the economy, with its huge deficits and mounting debt, depresses the private sector. Watching the central bank enable it all by purchasing some of the government’s debt is even more depressing."

http://tinyurl.com/24rhvooom

Public Library said...

Scott,

Great work. The Fed is doing the only thing it knows how. Provide more paper in times of economic stress, regardless if the economic stress was the result of dare I say it, too much paper!

The one black swan not priced into the market is bucking the mantra 'The Fed always gets what it wants'. You've quoted it here time and time again. However, I tend to beleive the Fed really has no better ability to persuade people then the next guy. We are talking human psychology, not economics.

Should the Fed miscalculate this fiat experiment, I shutter to contemplate the repercussions.

Benjamin Cole said...

Randy-

A cheaper dollar will help exports. Productivity has been doing excellently for years, and I think will only get better, thanks to a strong US venture capital community, and global transfer of even highly technical information instantly, on the web.
Aggregate demand is weak--since we do not like fiscal boosting, let's go to monetary. We can pull in our horns later, when the train is rolling again.

Indeed, if the Fed can show it can stimulate the economy through QE, the argument for fiscal stimulus will be weakened.

I see a better future on the fiscal front. We are getting out of Iraqistan, for beginners, and the public appetite for more debt seems limited.

I think we have a secular bull market ahead, if only the Fed gets some guts. No guts no glory.

theyenguy said...

I study the 30:10 Yield Curve Daily, $TYX:$TNX, and it today has completed its finaly steeping; it will be flatteinng; that is it will be going flat in months to come wiping out those invested in the Zeroes, ZROZ, and the 20 to 30 Year US Treasuries, TLT. The longer out debt is the worst palce to be.

The US Dollar, $USD, is oversold at 78.78 on 9-29-2010. Should the Euro FXE, go down, then stocks will go down. Better said, when the, EUR/JPY, that is the FXE:FXY, goes down, then the stocks will go down. Volatility pick up.

Banks KBE have fallen below support of 23. And European Financials, EUFN, have fallen from support of 22.5. Stocks could fall lower 9-30-2010 or 10-1-2010.

I am bearish stocks due to regional news coming out of Europe, such as the EU Banking Sector Stability Report for September 2010, produced by the ECB relates funding difficulties for a number of European banks.

I am bullish gold. As carry trade investment comes out of gold, it is likely to fall under $1,300 for a period of time, before it moves once again, substantially higher.

The Fed may announce QE 2 on November 3, 2010. If it does, it will be gold inflationary.

I had been thinking it wise to go short the market, as the EUR/JPY, has reached full expansion, but after reading Tyler Durden's article ... Why QE2 + QE Lite Mean The Fed Will Purchase Almost $3 Trillion In Treasurys And Set The Stage For The Monetary Endgame ... I've concluded that further fiat asset expansion is possible if the Fed does come out with a surprise QE 2.

Yes, QE 2 may come November 3, 2010, as the Federal Reserve may announce a plan to buy US Treasuries, it may simply print Dollars, yes simply print and print and print; and buy US Government Debt, to prevent a deflationary collapse.

This of course would send the value of the US Dollar, $USD, plummeting, and would be quite inflationary to many assets, such as food commodities, FUD, and especially gold, $GOLD. I think it wise to buy gold, specifically gold coins, at this time, even though a Surprise QE 2 is likely coming on November 3, 2010, which may create a demand for SHY and IEF.

theyenguy said...

I study the 30:10 Yield Curve Daily, $TYX:$TNX, and it today has completed its finaly steeping; it will be flatteinng; that is it will be going flat in months to come wiping out those invested in the Zeroes, ZROZ, and the 20 to 30 Year US Treasuries, TLT. The longer out debt is the worst palce to be.

The US Dollar, $USD, is oversold at 78.78 on 9-29-2010. Should the Euro FXE, go down, then stocks will go down. Better said, when the, EUR/JPY, that is the FXE:FXY, goes down, then the stocks will go down. Volatility pick up.

Banks KBE have fallen below support of 23. And European Financials, EUFN, have fallen from support of 22.5. Stocks could fall lower 9-30-2010 or 10-1-2010.

I am bearish stocks due to regional news coming out of Europe, such as the EU Banking Sector Stability Report for September 2010, produced by the ECB relates funding difficulties for a number of European banks.

I am bullish gold. As carry trade investment comes out of gold, it is likely to fall under $1,300 for a period of time, before it moves once again, substantially higher.

The Fed may announce QE 2 on November 3, 2010. If it does, it will be gold inflationary.

I had been thinking it wise to go short the market, as the EUR/JPY, has reached full expansion, but after reading Tyler Durden's article ... Why QE2 + QE Lite Mean The Fed Will Purchase Almost $3 Trillion In Treasurys And Set The Stage For The Monetary Endgame ... I've concluded that further fiat asset expansion is possible if the Fed does come out with a surprise QE 2.

Yes, QE 2 may come November 3, 2010, as the Federal Reserve may announce a plan to buy US Treasuries, it may simply print Dollars, yes simply print and print and print; and buy US Government Debt, to prevent a deflationary collapse.

This of course would send the value of the US Dollar, $USD, plummeting, and would be quite inflationary to many assets, such as food commodities, FUD, and especially gold, $GOLD. I think it wise to buy gold, specifically gold coins, at this time, even though a Surprise QE 2 is likely coming on November 3, 2010, which may create a demand for SHY and IEF.

Jason and Jen said...

Scott,

Do you understand what Brian Wesbury is saying in this article? His claim is that the Fed did not create bank reserves to fund asset purchases in QE1. Instead, he claims that bank customers simply wanted to hold money and put in on deposit - which then the bank's put into excess reserves - THEN the Fed purchased assets with this "borrowed" money...seems backward to me. What is your take?

http://www.forbes.com/2010/09/27/federal-reserve-economy-quantitative-easing-opinions-columnists-wesbury-stein.html

Scott Grannis said...

My take on QE is somewhat different from Brian's, but I don't think the differences are significant.

As I see it, the Fed purchased a bunch of assets and paid for them by crediting banks with newly created bank reserves. The banks haven't done very much with those reserves, however, leaving the vast majority of them on deposit with the Fed where they earn interest similar to what banks could earn with T-bills.

What this means is that the banking system apparently had a very strong desire to hold a less risky portfolio of assets. Banks are happy with tons of reserves on deposit at the Fed. Banks are not especially eager to use those reserves to fund new loans to customers, and customers are not too eager to get new bank loans. Most people are apparently still in deleveraging mode. All of this adds up to a banking system that exhibits a very strong desire for money (and very little desire to be short money).

When the demand for money surges as it has, it is appropriate for the Fed to accommodate that demand by increase the supply of bank reserves.

But things are not going to remain this way forever. If the demand for money declines and banks start using reserves to create new deposits, then things could get out of control quickly unless the Fed is equally quick to remove the reserves it has injected.

Plus, there are important signs that the Fed has over-supplied the system with money. The value of the dollar is weak and falling. Gold and commodity prices are rising. The yield curve is still quite steep, especially at the long end. Inflation expectations are rising.

Public Library said...

Here is a nifty chart of consumer prices going back to the early 1900's.

http://images.mises.org/4746/Figure2.png

It is obvious to see around Nixon's time when the money printing started in earnest.

To assume this is both needed and helpful as you describe is preposterous.

The Fed does not need to provide money when it is demand or reign it in when money is out of favor.

The Federal Reserve needs to stop playing with the money supply altogether before this shaky battleship sinks like the Bismark.

Jason and Jen said...

Scott,

I agree with all that you've written above on the mechanics and affects of QE. However, to say that you agree with the way Wesbury thinks of it doesn't make sense.

Specifically, he says that no new money was created by the Fed - via the creation of bank reserves - were created in order to buy the assets in the first place.

He thinks that QE was simply the Treasury issuing securities, depositing the money with the Fed and then the Fed purchasing assets from the private sector. No new money created. This is clearly wrong.

The Fed actually pushes buttons, creates new reserves, buys securities from the public, the public buys and sells assets for a while and then finally, the money is deposited with the banks.

The banks then lend a little here and there where their are opportunities, but this hasn't happened to the degree that the reserves are converted from "excess" to "required" reserves.

The money creation mechanism basically stalled out when the demand for credit and the bank capital levels are as weak as they are - a combination of the two. After this stall occurs, the bank reserves sit as excess reserves at the Fed.

Therefore, there is about $1 trillion sitting as excess reserves at the Fed...from the creation of reserves by the Fed.

This concept that people demanded money, liquidated securities, took the money and deposited it with their bank and then their bank lent it to the Fed...no new money created, just loans to the Fed from existing money already out there...is illogical, right?

If you will, tell me how your interpretation of QE is similar to Wesbury's when it comes to the mechanics...this "borrowed money" concept versus "printed money" idea.

Wesbury is saying that there is no new money in the "system" yet $1 trillion in excess reserves shows that somebody created it out of thin air.

Scott Grannis said...

Let me stipulate that the Fed bought $1 trillion worth of MBS; that this was "paid for" by creating $1 trillion of bank reserves; and that all of those reserves remain on deposit at the Fed as excess reserves. That's not all exactly correct, but it's a close approximation and simplifies the analysis.

Another way of describing what happened is that the Fed exchanged $1 trillion in bank reserves for $1 trillion of MBS. Since those bank reserves are functionally almost identical to T-bills (they pay about the same rate of interest and they are virtually default-free), we might say that this operation was the equivalent of the U.S. government selling T-bills to the public and using the proceeds to buy MBS. The Fed effectively "borrowed" money from the banking system at the T-bill rate and used the money to buy MBS. If Treasury had borrowed the money by selling T-bills, the net effect would have been the same.

Apparently the financial system on net had a desire to reduce its holdings of MBS (semi-risky securities) and increase its holdings of risk-free securities (T-bills or bank reserves). The Fed facilitated this trade.

The Friar said...

Scott,

I would be interested in having you out to speak at our University. If you are interested, please email me at eroot@westliberty.edu

W.E. Heasley said...

West Liberty?!? The oldest college in the great state of West Virginia!

Be sure to play the Palmer Course at Olgelby Park.

Jason and Jen said...

Scott,

I am sorry, but the analogy that the Fed's QE policy is just like the Treasury issuing t-bills and buying MBS isn't too clear to me.

In the case of the Treasury issuing t-bills, receiving $1 trillion in cash and then buying MBS with the cash, I would say nothing really happened. Cash was brought into the Treasury from one set of investors and sent back out to another set of investor when they bought the MBS. No new cash floating around at all...just different players with the cash.

In the case of the Fed's QE actions, the Fed issues bank reserves in exchange for MBS. My understanding is the bank reserves are actually immediately credited to the seller's bank's account with the Fed. In time, that original seller buys something with the cash and eventually the money lands in some other seller's bank account and it ends up back at Fed in the form of bank reserves on deposit - basically as excess reserves - all newly created, not provided by the public.

The t-bill issuance scenario uses existing money in the private sector to fund asset purchases. The newly created bank reserve issuance by the Fed come out of thin air, washes through the system, and is finally credited to the final receiving banking institution as excess reserves.

Just because excess bank reserves pay similar rates and have similar default protections as t-bills, I just don't see them being analogous in terms of "printing" vs. "borrowing".

Unless all your saying is the Treasury creates the borrowed money by issuing t-bills to the public and the Fed creates the borrowed money by printing it and having somebody end up with it in exchange for the asset they sold to the Fed.

It seems the big difference, however, has to do with the maturity structure of the government issued liability (t-bills versus bank reserves).

The Treasury's t-bills have a quick maturity with the government liability being destroyed quickly.

The Fed's bank reserves have an unknown and possibly very long maturity date and therefore may not be vanquished quickly.

I may not be thinking about this correctly - perhaps if, in my mind, I just collapsed the two structures (Treasury and the Fed) into one consolidated balance sheet, it would all become clearer to me.

Scott Grannis said...

J&J: Please, no need to apologize.

Part I of my response:

When Treasury wants to raise money to fund the government's deficit, it sells T-bills (and bonds, but let's just focus on bills) to the public. With the proceeds from the sale, it then pays the government's bills. Cash comes in from the sale of bills, and cash goes out to pay the bills. No new cash is created, as you point out, cash just changes hands a few times.

However, in the process of shuffling cash, the Treasury creates a new liability in the form of T-bills. This is a liability for the government and an asset for the private sector. (And by the way, the short maturity of the bill is irrelevant. It is only significant to the extent that short maturities insulate investors from interest rate risk, while they expose the government to interest rate risk.)

Now let's consider the Fed's QE program. The Fed's objective is to relieve the private sector of some significant quantity of longer-term bonds, and by so doing to increase the price of the bonds, thereby lowering long-term interest rates. This is presumed to stimulate economic activity—but that is another subject. In any event, an investor would describe this operation as borrowing short and investing long. It’s a form of leveraged arbitrage, since it involves no money up front but the payoff can be huge. The Fed profits on the spread between the bonds it buys and the interest it pays on reserves.

The Fed offers to buy bonds and accepts offers from large investment banks and institutional investors. It pays for the bonds by depositing reserves in the bank accounts of the sellers. The sellers presumably withdraw the funds from the banks and spend them on something else; for some period there are extra dollars floating around the banking system. The evidence to date points to those dollars coming back to the banking system in the form of deposits, never effectively leaving the banking system. We know that because excess bank reserves have increased by virtually the same amount as the Fed's purchases of bonds—the banks have effectively converted the new cash the Fed created into reserves. At the end of the day, the Fed creates no new money because the world has wanted to hold the extra money it received from the sale of bonds in the form of excess reserves. Excess reserves are in that sense equivalent to the bills that Treasury sells, since the reserves are a liability of the government and an asset of the private sector, and they pay an interest rate similar to that of bills.

So the Fed paid out cash to bond sellers, but that cash stayed in the banking system and banks elected to park that cash with the Fed in the form of reserves that pay interest, rather than lend that cash to some private party. We know that because there has been no increase in the money supply commensurate with the huge increase in reserves.

Think of it this way: the Fed pays out cash to bond sellers, then offers to sell reserves to those same bond sellers, and the sellers happily agree. Cash goes out and comes right back in. No new cash is created because the private sector is content to “buy” reserves with the cash it receives from bond sales.

Scott Grannis said...

Part II of my response to J&J:

The net effect of all this is that the Fed effectively swaps newly-created bank reserves for bonds. The Fed is left holding a new asset (bonds) and a new liability (reserves). No new money is created in the process. The Fed stands to profit on the spread between bond yields and the interest rate on reserves, and that profit, as Brian Wesbury has noted, comes at the expense of the private sector.

The difference between the Fed's QE and the Treasury's deficit financing operations is that the Fed ends up holding bonds that were formerly held by the private sector, whereas Treasury ends up holding nothing. But in both cases, the private sector (e.g., banks) ends up holding new assets in the form of bills and bank reserves. And in both cases, no new money (on balance) is created.

It doesn’t have to work this way, of course, since banks can use their reserves to create new deposits in the form of loans that pay a higher interest rate than reserves. With $1 trillion of reserves sitting idle at the Fed there is an astounding potential for monetary expansion. But so far, no one has been willing to do that trade. But they might in the future, and that is the thing that most concerns the world.

Jason and Jen said...

I am in total agreement with everything you just wrote. It was stated much more clearly than Wesbury, in my view.

The Fed's ability to squeeze their profitable arb play by raising the rate it chooses to pay on reserves does minimize the possibility of huge money creation. Doing so is costly to taxpayers on one hand, but it is beneficial to depositors as well in the form of higher income.

John said...

"People have been hoarding money since the financial crisis erupted, and the hoarding continues to this day. That has depressed growth in the economy, which is another way of saying that fear of the future and risk aversion are not compatible with healthy growth."

And herein lies the rub: on the street level, the most significant uncertainty relates to job security. If people don't feel secure in their jobs or confident about finding new jobs, they'll avoid further risk. Recent corporate policies, in good times and bad, have eroded job security.

Downsizing may increase productivity, and job insecurity. I'm not going to buy a house. The company I work for could be bought and dismantled tomorrow.