Five years ago, around the time that the visible signs of the current financial mess were becoming painfully obvious for even the most uninformed observer, the Federal Reserve held assets of something under $900 billion dollars. In the intervening time, the balance has grown to $2.8 trillion, with most of that growth occurring in the fall of 2008, and a second growth spurt occurring in the first half of 2011. This balance sheet depicts the Federal Reserve form of money printing, as the Fed has acquired assets using digital cash balances created from nothing.
The beneficiaries of these purchases are the system banks, receiving these digital balances. During this crisis, the banks have, for the most part, held these balances as excess reserves – reserves over and above the minimum necessary for regulatory requirements. The excess reserves, historically always at or near zero (meaning credit was provided to the maximum extent) have grown from this negligible amount to $1.6 trillion earlier this year, and now stand at $1.4 trillion.
Thus, out of a $1.9 trillion growth in the Fed balance sheet, $1.4 trillion is still held by the banks, with $500 billion lent out and circulating in the system.
I have speculated that the reason the banks hold excess reserves when this had not been the case previously is because, in fact, these aren’t excess. The banks needed the liquidity and perhaps the solvency. Underperforming assets (meaning little or no cash flow relative to balance sheet mark) are held by the banks at values above current market, therefore depicting an excess of assets when in reality this does not exist.
It appears that the banks are now growing more comfortable with lending these excess reserves, as recent trends show a decrease in these balances. In fact, since the Fed recently announced QE to infinity, it hasn’t been the Fed balance sheet that has grown, but instead a reduction of excess reserves. In the meantime, the Fed’s stated policy remains – that of increasing its purchases to the tune of up to $85 billion per month of treasury and agency securities, although it remains to be seen if they act on this policy.
All of this is background for the issue of withdrawing, one day, the monetary stimulus (money-printing) provided by the Fed over this time. This issue has been raised by inflation hawks from the beginning of this Fed experiment, and was addressed formally by the Fed at times in the past.
A recent story in Bloomberg addresses this issue:
A decision by the Federal Reserve to expand its bond buying next week is likely to prompt policy makers to rewrite their 18-month-old blueprint for an exit from record monetary stimulus.
Not only has the Fed not removed the increase of $1.9 trillion, it is planning further asset purchases thus further compounding the problem of removing this stimulus in the future. Further, as banks reduce the excess balances by lending these out (to the tune of $500 billion so far), removing the stimulus becomes exponentially more complicated.
“There is certainly an issue about unwinding the balance sheet” in a way that “is effective and continues to support the recovery without creating inflation,” St. Louis Fed Bank President James Bullard said in an interview in October.
There are two issues, 1) the stimulus provided that is held as excess reserves, and 2) the stimulus provided that has been lent out into the economy. Neither can be unwound without slowing down the economy, but the second would be especially problematic.
The Fed can reduce its balance sheet by selling securities into the market, and retiring the acquired digits to the same place from which they came – into thin air. To the extent these securities represent excess reserve balances, this will not directly harm bank credit provided to customers.
However, any such selling program by the Fed will drive prices of the securities down and drive interest rates up, thus removing support from the economy. Further, removing the stimulus that has found its way into bank lending will directly shrink credit provided to the general market, also removing support from the economy.
As to unwinding without creating inflation, I don’t get this statement by Bullard. Everything I understand about monetary policy and economics suggests that it is the risk of deflation that stands in the way of such a policy of unwinding.
The goal is to return the balance sheet to a pre-crisis size in two to three years and eliminate holdings of housing debt “over a period of three to five years.”
This grows ever more difficult if the Fed finally makes good on its QE to infinity pledge, and compounds as the banks reduce their excess reserve balance via lending into the market. There is direct evidence of the latter, and so far only lip-service to the former.
“The exit is going to take a long time,” said Stephen Oliner, a resident scholar at the American Enterprise Institute in Washington and former Fed Board senior adviser.
Not just a long time. The Fed is five years into this gamble-the-civilized-world-and-potentially-destroy-the-division-of-labor experiment, and so far every indication and action continues to move against an exit.
“We are deep into experimentation at this point,” Oliner said. “It’s understandable that people are worried.”
It is quackery. They don’t know what they are doing and they have even less idea of how to get out of the mess they made.