Jeff Harding

Bernanke’s end game

Disminuir tamaño de fuente Aumentar tamaño de fuente Texto Imprimir esta página

 

 

Daily Capitalist logo

I follow Rick Davis’s data from Consumer Metrics Institute and he has come out with several interesting (and proprietary) ways to analyze the economy by measuring consumer data. His latest report shows a continuing “contraction” in consumer data. He describes it as follows:

“Contraction Watch” follows [our] Daily Growth Index and compares it on a day-by-day basis with the same index during the dip created by the “Great Recession” (with the plotting for each event starting on the first day that the Daily Growth Index started to contract). …

[T]he Daily Growth Index is rapidly approaching its previous record low value set on October 4, 2010 — and it has again surpassed the lowest values seen during the 2008 contraction event. Regardless of what is happening in the economy as a whole, the consumers that we track are either not participating in the upside or have chosen to be cautiously frugal until they see clearer signs of recovery.

What he is saying is that this thing is lasting far longer than anyone anticipated. It fits in with my “two-tier” economy idea where the investors are benefiting from QE but it hasn’t trickled down, yet.

Graph depicting Consumer Metrics Institute's Contraction Watch for 04-12-2011

Rick also wrote an essay on quantitative easing that I agree with, and it backs up some of the things I have been saying for some time. I could quibble about his remarks about China and its dollar holding, but this is an excellent piece. I think it’s a pretty good bet that we’ll see QE 3 if unemployment stays high. I am republishing his piece with his permission.

QE-2 End-Games – Mr. Bernanke’s Dilemma

“Consider for a moment that there are only three broad options for Mr. Bernanke at the end of June:

– Unwind the cumulative effects of QE-1 and QE-2, sending us back to late 2008 in more ways than we would care to remember;

– Restrain new purchases but hold the current inflated balance sheet levels intact, eventually causing budget-busting debt servicing costs as the Treasury falls into the unwelcoming arms of the “bond vigilantes”;

– Continue QE either directly or indirectly, while selling “Main Street” on how great the equity markets have done.”

QE-2 is supposed to wrap up sometime in June. The question is: will it actually end in June? Or, to rephrase that question in a more political context: is it in anybody’s interest to have it end then?

It may be useful to reflect back to a time a little over two years ago when most economists were concerned that the Federal Reserve had expended all of its monetary policy weaponry against the “Great Recession” to no avail, caught in a version of what Keynes characterized as a “liquidity trap”. Interest rates were effectively zero, and yet credit was not growing and the economy was not recovering. It was becoming clear that two factors were at work:

– Banks were loathe to loan new monies to non-credit-worthy borrowers;

– And other borrowers who were still credit-worthy simply wanted no more debt (explaining in large part why they had remained credit-worthy in the first place).

In fact, even existing credit was being extinguished as credit-worthy borrowers deleveraged and non-credit-worthy borrowers defaulted. In short, we were experiencing a very real credit collapse. And the Federal Reserve was then presumed to have no conventional monetary tools left that could reverse that situation.

That’s when Mr. Bernanke entered the realm of the “unconventional,” borrowing a tool previously used by the Bank of Japan to deal with its own great recession — now nearly two decades old. That tool was “Quantitative Easing” (QE) which amounts to the central bank injecting liquidity directly into the banking system by swapping fabricated-from-thin-air excess bank reserves for outstanding governmental or commercial debt instruments. From a balance sheet standpoint this process essentially expands the Federal Reserve’s assets by the amount of the purchased debt instruments and offsets those assets with liabilities in the form of freshly minted bank reserves electronically fabricated and credited to the money-center banks’ accounts.

The Federal Reserve’s initial experiment in QE (starting in November 2008 and now known as QE-1) was also used to move vast amounts of lower quality mortgage-backed securities from the balance sheets of commercial banks to the balance sheet of the Federal Reserve. By June 2010 the Federal Reserve had increased its total portfolio of securities by $1.58 trillion, of which $1.13 trillion was mortgage backed securities. In the first quarter of 2010 it was thought that a full recovery was underway, and the Federal Reserve began to contemplating “unwinding” its positions.

But by August 2010 the recovery had begun to look less robust, and another round of QE (now known as QE-2) was announced. The initial QE-2 purchases were designed to simply offset the shrinking value of the mortgage-backed securities with fresh U.S. Treasury issues (because the value of the Fed’s mortgage-backed securities portfolio has shrunk by nearly 18% since May 2010 as a consequence of normal principal payments, refinancing retirements and some foreclosure activity). But by December 2010 the purchases had increased substantially beyond what was necessary keep the total portfolio balance level, bringing the total of all Federal Reserve QE purchases (QE-1 and QE-2) through early April 2011 to $2.02 trillion, with an additional $0.20 trillion planned before the end of June 2011.

It’s reasonable to ask what over $2 trillion of additions to the Federal Reserve’s balance sheet has accomplished:

– Since QE-1 commenced in November 2008 the Federal Reserve’s Employment-Population Ratio has dropped from 61.4% to 58.5% (i.e., nearly 3% less of the population is employed). Meanwhile the official headline unemployment rate grew from 6.8% to 8.8%;

– Household income has stagnated in the face of price inflation in household staple commodities: corn prices are up 103%, oil prices rose 122% and cotton spot prices quadrupled;

– The longer end of the Treasury’s yield curve rose over 200 basis points (80%) from a bottom reached in December 2008.

– On the home front, residential real estate prices have dropped an additional 9%, housing starts have dropped an additional 10%, and foreclosure filings increased 23%;

But on the brighter side of things:

– The price of gold is up 75%, the S&P 500 is up 48% and earnings per share for the S&P 500 are up 39%.

As with many experiments, QE has generated some unexpected results. As an example, by conventional economic reasoning the injection of vast amounts of new liquidity should not have driven long bond rates (or mortgage rates) up. As another example, banks were expected to put more of that liquidity to use in ways that stimulated real commerce.

Looking at the big picture, we might wonder: what are the observable macro-economic results of the Federal Reserve’s great QE experiment since November 2008? At least two observations come to mind:

– There has been a bubble in financial assets and commodity prices, even as real estate prices have continued to deflate. This has been largely due to the mechanism used for the injection of the liquidity — increasing the excess bank reserves of money-center banks. Those banks, in turn, deployed the excess reserves in the ways that they felt would generate (for them) the greatest profits — speculating on financial assets and commodities (and certainly not real estate or mortgages).

– Despite the financial bubble, “Main Street” Americans have fewer jobs per capita, lower household worth and less disposable income. This can be considered collateral damage from the use of self-serving banks as the liquidity recipients, since essentially none of the liquidity has been put into productive capital that can grow U.S. commerce, jobs or median real household income.

The experience of “Main Street” Americans during QE has not been materially different from the experiences of their Japanese counterparts over the past ten years — a history that the Federal Reserve surely understood before injecting $2 trillion of liquidity onto the balance sheets of select U.S. banks. So, why has QE become the stimulus tool of choice?

The Bank of Japan’s answer is simple: presumably things would have been much worse without QE — i.e., the injected liquidity neutralized (to some extent) an ongoing deflationary credit contraction. And, not so incidentally, they had no other tools left in their monetary toolkit.

By the way, after nearly ten years of QE, the Bank of Japan has not been able to execute an exit strategy. But they have learned two critical things:

– Nobody really cares about central bank balance sheet bloat; and,

– They were able to monetize and even further domesticate their government’s sovereign debt, keeping the unprecedented debt/GDP ratio affordable and deferring the need to address the structural fiscal deficit issues — and all without trashing their currency’s exchange rate.

As a stimulant for Japan’s economy, QE was arguably a complete failure. But by happy accident, the experiment managed to kick an a 200% debt/GDP ratio (twice as bad as in the U.S.) down the road for a decade.

That happy accident is not trivial, nor is it likely to have been overlooked by Mr. Bernanke. As little as a year ago economists were concerned about the possibility of the U.S. Treasury’s largest creditor (China) suddenly dumping their holdings to repatriate their stash of foreign exchange surpluses for domestic use. That is literally no longer a concern — simply because the Treasury’s largest creditor is now the Federal Reserve (compare the Fed’s current $1.35 trillion holdings to China’s measly $1.16 trillion). The Fed has roughly doubled its holdings over the past year, and at their current pace of year-over-year Treasury acquisitions Mr. Bernanke could (and almost certainly would) sop up the entire Chinese position in less than two years.

Perhaps more importantly, for the past five months the Federal Reserve has bought and monetized the entire net new U.S. Treasury debt. If we engage in some (presumably) hypothetical arithmetic in something akin to economic “magical realism,” the Federal Reserve has put into place and tested a means of monetizing over $100 billion in new U.S. Treasury debt per month — potentially (and magically) financing a $1.4 trillion annual fiscal deficit for at least a decade (if Japan’s experience and 200% debt/GDP ratio are any guide). And above all, on the short term this financing would be painless for the electorate and a bonanza for the financial elite.

Consider for a moment that there are only three broad options for Mr. Bernanke at the end of June:

– Unwind the cumulative effects of QE-1 and QE-2, sending us back to late 2008 in more ways than we would care to remember;

– Restrain new purchases but hold the current inflated balance sheet levels intact, eventually causing budget-busting debt servicing costs as the Treasury falls into the unwelcoming arms of the “bond vigilantes”;

– Continue QE either directly or indirectly, while selling “Main Street” on how great the equity markets have done.

If Mr. Bernanke has learned his lessons well, there is no dilemma. It’s an almost magical “no-brainer” that only our children may regret.

Copyright © 2011 · The Daily Capitalist

* * *

Published by kind permission of Jeff Harding.
Link to original version

 

Comparte este artículo

0 comments