Enquiring Minds Want to Know: Is David Malpass Correct about Fed Tapering?
by Paul Kasriel, The Econtrarian
Back on March 13, David Malpass, president of Encima Global LLC, wrote an op-ed for The Wall Street Journal entitled “The Fed’s Taper Is Already Paying Off”. In this op-ed, Malpass argues that the Fed’s purchases of securities, aka Quantitative Easing (QE), have inhibited bank lending to small businesses, aka job creators.
Malpass has observed that with the Fed’s commencement of the tapering of its QE in January and February of this year, bank lending has accelerated. Malpass argues that the slowing in QE is the reason bank lending has accelerated. Malpass concludes that the Fed’s tapering of its securities purchases is, in and of itself, bullish for the US economy.
Having mellowed (decayed?) since becoming eligible for Medicare, I now normally would read an op-ed like that of Malpass, chuckle, and then go practice my bass guitar. As a matter of fact, I cancelled my subscription of The Wall Street Journal several years ago because of the consistently poor analytical quality of its economic-related op-eds. So, I would not have run across Malpass’s (Strunk says to add the “s” after the apostrophe) op-ed if it were not for some of the readers of my doggerel calling my attention to Malpass’s op-ed. Given that I have argued in recent commentaries that, all else the same, Fed tapering is bearish for nominal US economic activity and the behavior of risk-asset prices, some of my readers asked my opinion of Malpass’s thesis. Well, if enquiring minds want to know, here goes.
“Rather than creating or printing money, as is often assumed, the Fed borrows heavily from banks to buy bonds … In 2013 alone, the central bank borrowed nearly $1 trillion from the banking system. It wasn’t created out of thin air. It is recorded as a liability of the Fed and an asset for banks… The Fed didn’t create any extra currency or private bank deposits. To make their loans to the Fed, banks had to reduce other assets.” [emphasis added]
These comments by Malpass negate an important lecture given in every Money and Banking 101 course. That lecture explains how a central bank can stimulate nominal spending in an economy by purchasing securities in the open market. In a monetary system in which banks are required to hold only a fraction of their deposits as cash reserves with the central bank, the banking system is able to generate new amounts of credit and deposits that are some multiple of the dollar amount of securities purchased by the central bank. Ever since Professor Chester A. Phillips articulated the deposit and bank credit multiplier back in 1920, enrollees in most Money and Banking 101 classes have been taught that central bank purchases of securities stimulate bank credit (the asset side of the banking system’s balance sheet) and bank deposit (the liability side of the banking system’s balance sheet) expansion . So, Malpass’s argument that Fed purchases of securities “didn’t create any extra currency or private deposits … and [t]o make loans to the Fed, banks had to reduce other assets” represents a new theory of money and banking that certainly deserves critical scrutiny. So let’s scrutinize.
Let’s go through the accounting of a Fed purchase of securities. (This would be a less tedious exercise if I knew how to draw T-accounts in Word, but I don’t. So, if you are interested in reviewing the lecture in Money and Banking 101, brew a pot of coffee before going on.) The Fed conducts its securities purchases and sales with certain designated (primary) government securities dealers. These dealers are essentially conduits between the Fed and ultimate sellers/purchasers of Treasury and Agency securities. If the Fed solicits offers on securities it wants to purchase, these primary dealers, in turn, contact their customers – e.g., banks, insurance companies, pension funds, mutual funds – with bids to purchase securities from them.
Assume that Primary Dealer A purchases $100 of securities from Pension Fund B and simultaneously sells these $100 of securities to the Fed. Let’s follow the “money”. The Fed credits Primary Dealer A’s bank account at Bank A for $100. On the books of Bank A, the cash asset item “deposits at the Fed” goes up by $100 and the liability item “deposits payable to Primary Dealer A” goes up by $100. On the books of the Fed, the asset item “securities owned outright” goes up by $100 and the liability item “deposits payable to Bank A” goes up by $100.
Let’s pause here to contemplate what has taken place. The Fed has purchased $100 of securities from Primary Dealer A and Primary Dealer A now has $100 more in deposits (and $100 less in securities) than it had before the transaction with the Fed. From whence did this $100 increase in Primary Dealer A’s deposits come? From me? From you? No, from the Fed. And from whence did the Fed get these deposits? It created them figuratively “out of thin air”.
Okay, let’s get back to the story. Remember that Primary Dealer A simultaneously purchased $100 of securities from Pension Fund B while selling $100 securities to the Fed. Pension Fund B needs to be paid for the $100 of securities. Primary Dealer A, immediately upon notice that its deposits at Bank A have been increased by the Fed, orders Bank A to transfer $100 to the deposit account of Pension Fund B at Bank B. On the books of Bank B, the cash item “deposits at the Fed” goes up by $100 and the liability item “deposits payable to Pension Fund B” goes up by $100. From whence did the $100 increase in Pension Fund B’s deposits come? From the deposits of Primary Dealer A. From whence did the $100 of deposits of Primary Dealer A come? From the Fed. And from whence did the $100 from the Fed come? From thin air.
Alright, let’s complete the bookkeeping. With the payment for securities to Pension Fund B, on the books of Bank A, the cash item “deposits at the Fed” goes down by $100 and the liability item “deposits payable to Primary Dealer A” goes down by $100. The only change on the books of the Fed is that the liability item “deposits payable to Bank A” changes to “deposits payable to Bank B”.
Pension Fund B and the economy as a whole now have $100 fewer securities and $100 more deposits. More than likely, but not definitively, Pension Fund B will want to purchase some new earning assets to replace the $100 of securities it sold via Primary Dealer A to the Fed. Given that the Fed has bid down the yield on government/agency securities as a result of its QE purchase of such, maybe the pension fund will want to replace the $100 of securities it sold with some higher-yielding (and, thus, riskier) securities – say corporate bonds or securities backed by auto loans. Whatever securities purchased by Pension Fund B to replace those ultimately sold to the Fed, Pension Fund B will be paying for its new securities acquisition with funds ultimately created out of thin air. This means that the Fed’s securities purchase has added $100 of net new credit to the economy, credit created out of thin air. And that means that the recipients of this net new credit can increase their current spending on something – a good, a service, an asset – while no other entity need pare back its current spending.
If, for some reason, Pension Fund B chose not to replace the $100 of securities it ultimately sold to the Fed with acquisitions of new securities /loans, then credit to the economy would not decline, but rather, would be unchanged. The sale of securities to the Fed does not reduce credit to the economy. It simply changes the name of the creditor from Pension Fund B to the Fed. No credit has been withdrawn as a result of this. So, at worst, Fed purchases of securities in the open market do no harm with respect to credit creation for the economy. If Pension Fund B chooses to replace some or all of the securities it ultimately sold to the Fed with the acquisition of new securities/loans, then Fed purchases of securities will have resulted in a net increase in thin-air credit to the economy.
Now let’s see what the net effect of the Fed’s $100 purchase of securities has had on the balance sheet of the banking system. For the banking system, the cash asset item “deposits at the Fed” (aka, bank reserves at the Fed) increased by $100. For the banking system, the liability item “deposits” has increased by $100. What has happened to bank loans and securities, i.e., bank credit? Nada. What has happened to total bank financial assets — deposits at the Fed plus bank credit? Total bank assets increased by $100. So, for the Fed to increase its securities holdings, banks did not need to reduce their securities holding or their loans.
In a fractional required reserve monetary system like ours and almost every economy’s, the banking system, which is required to hold only some fraction of the extra $100 of deposits it now has on its books as deposits at the Fed (cash reserves), could, in theory, expand its holdings of loans/securities by some multiple of the extra $100 of deposits at the Fed (cash reserves) it has on its books – deposits at the Fed created by the Fed out of thin air. Capital and/or regulatory constraints could prevent the banking system from further increases in its loans/securities. But even if this were the case, the Fed’s purchase of $100 of securities still would have resulted in a $100 net increase in thin-air credit to the economy (assuming Pension Fund B replaced the securities it ultimately sold to the Fed with others) and total banking system assets would have increased by $100.
Now let’s look at a slightly different scenario. Instead of Primary Dealer A acquiring $100 of securities from Pension Fund B in order to sell to the Fed, assume that Primary Dealer A acquires $100 of securities from Bank B. The net effect of this is that Bank B’s asset item “securities” decreases by $100 and its asset item “deposits at the Fed” increases by $100. Thus, for Bank B and the banking system at this point, total financial assets are unchanged. The banking system did decrease its holdings of securities in this case, but it ultimately was not required to, as will be explained in the next paragraph. For the banking system in this case, loans and securities would have decreased by $100 while total financial assets would have been unchanged because the asset item “deposits at the Fed would have increased by $100, countering the $100 decrease in the asset item “loans and securities”.
Bank B, in theory, could have replaced the $100 of securities it sold to the Fed via Primary Dealer A with the purchase of some other securities from some nonbank entity or the granting of some new loans to a nonbank entity, just as Pension Fund B was assumed to do in the first scenario. If Bank B did choose to replace the $100 of securities it ultimately sold to the Fed, then for Bank B and the banking system, total financial assets would have increased by $100, with deposits at the Fed increasing by $100 and loans/securities on the books of Bank B and the banking system having a net change of zero – Bank B sold $100 to the Fed via Dealer A and then acquired $100 of new securities and/or loans.
Why might not Bank B have chosen to replace the $100 of securities it ultimately sold to the Fed with the acquisition of some new securities and/or loans? For starters, Bank B can earn 0.25% annualized on overnight deposits held at the Fed. So, if the $100 of securities it ultimately sold to the Fed did not earn much above 0.25%, then it would have little incentive to replace them. One problem with this explanation is that the Fed has not been purchasing short-maturity Treasury/agency securities that might have a yield close to 0.25%. Rather, the Fed has been purchasing longer-maturity Treasury/Agency securities with yields considerably higher than 0.25%. In 2013, the yield on the Treasury 10-year security averaged 2.35%. (As an aside, banks typically do not hold longer-maturity Treasury and Agency securities because of the interest rate risk involved in such portfolio items.)
A better explanation as to why Bank B might not replace the $100 of securities it ultimately sold to the Fed is that Bank B might be constrained by its capital in increasing its assets. Regulators have increased required capital ratios on banks since the failure of Lehman Brothers in 2008. Regulators are restricting banks’ decisions to pay dividends or buy back their own equities depending on the capital positions of banks. If, in fact, banks have been reluctant to replace securities ultimately sold to the Fed with new acquisitions of securities/loans, it is not because banks had to reduce securities/loans “to make room” for “loans” to the Fed. In other words, the failure of banks to accelerate the growth in their loans and securities in 2013 had nothing to do with the Fed’s QE program. More likely, it had much to do with “tighter” regulatory issues. And that, precisely, is the rationale for QE. If banks are unable to increase their loans and securities because of capital and/or regulatory constraints, the Fed can temporarily increase its credit provision to the economy as banks repair their balance sheets.
Let’s look at some data. Let’s try to get an idea of exactly from whom the Fed has been purchasing Treasury and Agency securities – from banks or from others? Chart 1 shows the Q4/Q4 dollar change in Treasury/Agency securities holdings of the Fed, depository institutions (primarily banks but also S&Ls and credit unions), and all other holders of these securities, including foreign entities. In 2009, when QE I was in full swing, holdings of Treasury/Agency securities by depository institutions actually increased from 2008. In contrast, holdings of Treasury/Agency securities by all other sectors contracted. So, during QE I, the net sellers of Treasury/Agency securities were not banks, S&Ls and credit unions and, obviously the Fed, but everyone else. During QE II in 2011, everyone was a net buyer of Treasury/Agency securities, but depository institutions increased their holdings by less than they did in 2010. Now we come to QE III in 2013, the main focus of Malpass’s op-ed. As during QE I, during QE III, the net sellers of Treasury/Agency securities were everyone other than the Fed and depository institutions. It should be noted that depository institutions increased their holdings of these securities by less in 2013 than they did in 2012, but they were not net sellers. Notice also that net annual acquisitions of Treasury/Agency securities by depository institutions have been trending down since 2009, QE or not.
Let’s get a bit more granular by looking at the month-to-month dollar changes in commercial bank holdings of Treasury/Agency securities and dollar changes in their holdings of loans, leases and securities other than Treasuries and Agencies. These data for 2013 are plotted in Chart 2. Notice that in each of the last three months of 2013, when Fed tapering had not yet commenced, bank holdings of both Treasury/Agency securities and the aggregate of all other elements of bank credit – loans, leases and “other” securities – increased. If Fed open-market purchases of Treasury and Agency securities necessitated banks to reduce other assets, as Malpass asserts, how would he explain the fact that in the last three months of 2013 bank holdings of Treasury and Agency securities, the very securities that the Fed was purchasing, increased along with increases in all other assets included in bank credit?
“In 2013, the banking system’s total assets grew less than $900 billion, requiring roughly $100 billion in reductions in other types of bank assets to make room for loans to the Fed.”
Actually, in 2013, a full year of no-tapering QE, the banking system’s total assets grew by $934 billion on a Q4/Q4 basis. But let’s not quibble over a few billions of dollars. What Malpass failed to tell you but what you can see in Chart 3 is that 2013 marked the largest Q4/Q4 dollar increase in commercial bank total assets since 2008:Q4! Also, commercial bank holdings of loans, leases and securities other than Treasuries and Agencies increased by $217 billion. Admittedly, the 2013 increase in commercial bank holdings of loans, leases and “other” assets was $17 billion less than the 2012 increase, but I fail to see any data that would support Malpass’s claim that banks reduced any assets “to make room for loans to the Fed”.
“The data are clear that credit has been channeled away from growth, in effect rationed to the safest and best-connected creditors (government and big business) at the expense of those more likely to create jobs.” [emphasis added]
Did Fed securities purchases result in “channeling away” credit from small businesses or was it banks’ inability to lend to small businesses because of capital and regulatory constraints? Perhaps the Fed’s quarterly Senior Loan Officer Opinion Survey on Loan Practices might shine some light on this issue. Chart 4 shows the results of these surveys with regard to banks’ lending terms to small businesses. As the last recession took hold, the percentage of bank respondents in the survey tightening their lending terms jumped from 20% in 2008:Q1 to 60% in 2008:Q4. By the end of 2011, the percentage of bank respondents who were tightening lending terms to small businesses had drifted down to zero or close to zero. If only a very small percentage of bank respondents were reporting a tightening in lending terms, does this necessarily imply that a correspondingly large percentage were easing their lending terms to small firms? It ain’t necessarily so. A large percentage of bank respondents tightened their lending terms during the last recession and although not tightening those terms further, they might have just kept them “bar tight” thereafter. This is what appears to have happened. As shown in Chart 4, as the percentage of bank respondents reporting tightening terms began to decline, the percentage reporting unchanged began to ascend. The percentage of bank respondents reporting an actual easing in lending terms has moved up, especially in the second half of 2012, but in absolute terms, remains low.
So, I would argue that a logical explanation for the reason that less bank credit has been “channeled” to small businesses is that because of the inherently larger credit risk involved with of small business loans in an era in which bank regulators have a diminished tolerance for bank risk, banks have chosen to channel relatively little credit toward small businesses. Logically, there is nothing in Fed securities purchases that would preclude banks from stepping up their lending to small businesses if they chose to.
Although Malpass’s argument that Fed purchases of Treasury and Agency securities has “channeled” credit away from job creators is fallacious, I would posit that if the Fed had lent directly to small businesses rather than purchasing Treasury/Agency securities, this would have had a more stimulative effect on nominal goods/services spending. But, of course, this would put the Fed in the position of “picking winners and losers”, something Malpass abhors. And forget about Malpass abhorring it. Senator Rand Paul would have become apoplectic!
I would like to close this section with a criticism – not of David Malpass, but of the staff of the editorial page of The Wall Street Journal. Back in the days when Lindley Clark was associated with The Wall Street Journal, an article as devoid of logic and unsupported by data as this one by Malpass doubtfully would ever have been printed. There was a time in my life when the editorial page of The Wall Street Journal was my first read of the day. The op-eds related to economics were logical and supported by facts. Sadly, in my opinion, the quality of The Journal’s economic op-eds deteriorated significantly sometime not long after Lindley Clark’s retirement in 1995. As I mentioned at the outset, I canceled my subscription to The Wall Street Journal several years ago and now my first read of the day is The Financial Times. After reading Malpass’s op-ed, I am even more confident in the correctness of my decision.
And March still was much colder than normal. The blast of thin-air credit that has been created in recent months – not for the reason Malpass has given – ought to really propel economic activity in the second quarter when temperatures start to rise!
And Now a Sad Note
My 1995 Subaru went to donation heaven a few weeks ago. I guess it was time. For four years, the only thing illuminating the interior at night was the “check engine” light for some known minor issue. Several weeks ago, the “Stink-Mobile”, as my family affectionately referred to Old 95, was sideswiped by a pickup truck with a snowplow while I was at band practice. The next week, as I was driving Old 95 to band practice, suddenly, another light brought illumination to the interior – the “oil pressure” warning light. That marked a repair too far. Perhaps I have not learned my lesson, but despite the fact that the old Subaru gave me only 18-3/4 years of service, I opted to buy a new Subaru, thus contributing to that March spike in new light vehicle sales. Another lesson not learned – I still am going to band practice!
The views expressed above solely reflect those of Econtrarian, LLC.