In response to the 2007-2008 financial crisis and the ensuing Great Recession, some central banks, including the U.S. Federal Reserve, put some unconventional monetary policies in place.  Among these policies were large-scale asset purchase programs (LSAPs) which are more commonly referred to in the U.S. as quantitative easing (QE) measures. The Fed put three waves of QE in place.

Use of these measures has led to debates about their effects on economic outcomes in general and especially financial stability, with proponents emphasizing that they increased confidence and risk taking and therefore stimulated overall economic activity and sped up the recovery.  Critics argue these policies were ineffective or inefficient and that they may have set the stage for the next financial crisis by encouraging excessive risk shifting, fueling asset bubbles and creating incentives to "reach for yield."

Three Federal Reserve analysts, Robert Kurtzman, Stephan Luck, and Tom Zimmermann have just released a discussion paper titled "Did QE lead banks to relax their lending standards? Evidence from the Federal Reserve's LSAPs."  The paper is the results of research on how QE affected lending standards, looking specifically at the first and third rounds, QE1 and QE3 when the Fed purchased both MBS and Treasury Notes and using QE2, when only Treasuries were involved, as a control.

The research uses a standard, public set of bank balance sheet data and two confidential surveys that measure bank risk taking; The Federal Reserve's Senior Loan Officer Option Survey on Bank Lending Practices (SLOOS), conducted quarterly at the bank lending officer level, and Survey of Terms of Business Lending (STBL) that provides loan level data on newly issued loans to businesses and contains, among other variables, information on loan size, spreads, and internal risk ratings.

The study utilized SLOOS information collected from 2007 to 2014 and represents 50 percent of total assets of all commercial banks. Data was taken from questions about changes in standards and information on general loan demand.

The STBL data covered 2007 through 2014 and represented around 60 percent of all assets of U.S. commercial banks.  While SLOSS data was primarily from large banks, STBL information was from banks of all sizes.

The research looked at the variation in holdings of mortgage-backed securities (MBS) based on the share of those holdings in overall portfolios. Prior to QE1, the difference between the ratio of MBS to total securities between a bank at the 75th percentile of this share distribution and one at the 25th percentile was around 30 percentage points. Despite this substantial variation in MBS holdings, the banks in the main study sample were otherwise similar in terms of other balance-sheet characteristics.

The researchers employed a difference-in-differences design, looking at the average MBS ratio prior to a round of QE and then at the extent to which those banks with a larger share of MBS changed their lending standards and internal risk ratings for new loans after a round of QE. 

The identification rests on the idea that the MBS portfolios of banks are affected differentially in response to Fed's MBS asset purchases. Their identification is supported by the observation that, although the QE programs affected a broad range of asset prices in a similar manner to conventional policy, evidence suggests that QE operated through a "narrow channel.  Most relevant, QE policies affected the liquidity and returns of the assets targeted in the purchases more than other assets.

The research provided evidence that the first and third rounds of QE led to a lowering of lending standards and increased risk taking among those banks with the higher levels of MBS on their books.  In particular, it found that not only did QE induce banks to lend more, but to reshuffle their lending toward riskier loans. "We find that banks with more MBS holdings chose less tight lending standards after QE1 than banks with low MBS holdings, and that banks with more MBS holdings were more likely to ease their lending standards compared to unaffected banks after QE3. Moreover, consistent with our identification strategy, we find that there was no change in lending standards and risk-taking on new loans across bank MBS holdings after QE2, during which the Federal Reserve bought exclusively Treasuries," the authors say.

The effects of MBA purchases are about the same magnitude for both QE1 and QE3.  The lending standards index, (with a range of -1 to 1) of a bank at the 75th percentile of MBS ratio was 0.07 units lower after the two QE tranches than the index of a bank in the 25th percentile.  Similarly, the average internal risk rating of new loans (which can take value from 1 to 5, with 1 being low risk) originated by a bank in the 75th percentile was 0.05 units higher than those at a bank in the 25th. The effect is roughly comparable to the effect, cited in the literature, of a one percentage point decrease in the Fed funds target rate during times of conventional monetary policy.

There is a question as to whether results are driven by an increased supply of risky loans or by a demand for them.  The SLOOS data allowed controlling for demand as expressed by loan officers. There was a correlation between lending standards and perceived demand, but the results were not altered by controlling for demand.

The SLOOS data allowed researchers to not only examine the aggregate standards and loan demands of individual banks but also to examine them by loan category.  Thus, they could determine that, in QE1, the lowering of lending standards was mainly through less tightening of real estate loans, but in QE3 the effect came mainly from easing standards on commercial and industrial loans. 

Furthermore, the SLOOS data also allowed tracking the reasons provided by the banks for the changes in lending standards. The researchers tested whether increased risk-taking was driven by an improved capital position, a net-worth channel or increased liquidity for MBS that facilitates loan portfolio re-balancing, a "liquidity channel." The survey evidence indicates that the effect during QE1 results mostly from an improved capital position and from increased risk appetite, consistent with the "net-worth channel." Furthermore, the balance sheet evidence indicates that banks with more MBS have higher realized and unrealized gains after QE1.

The research did not allow a conclusive determination of what drove risk-taking during QE3; it does not appear to be either an improved liquidity position, economic outlook, or capital position.  The research also looked at whether the tapering of QE3 had a significant effect on bank risk-taking which was signaled over the course of 2013 and implemented in January 2014. Here too, banks with higher MBS shares tended to tighten their standards more/ ease less after the tapering had been implemented.

The authors ran several robustness checks and found:

  • Their results were not driven by an expansion of lending.
  • The purchase of Treasuries did not have an effect on lending standards and risk-taking during any round of QE.
  • There were no effects under various alternative definitions of their MBS measure, including varying the period over which pre-QE MBS holdings are calculated, eliminating the concern that some banks systematically bought MBS in anticipation of QE.

In summary, the study documents a causal effect of the Federal Reserve's LSAPs on risk characteristics of newly issued bank loans, with the first and third round having a significant effect on bank lending standards, roughly similar to a one percentage points decrease of the Fed fund target rate during normal times.  The authors call their results robust along a number of dimensions, including changes to the definition of the exposure variable, and assumptions about the timing of LSAPs.  The results therefor show that QE not only affected the availability of credit but also led banks to issue relatively more risky loans with lower lending standards.

The authors conclude that while their paper can speak to the causal effect of QE on bank lending standards and bank risk-taking, it does not allow an inference about whether the additional risk-taking was desirable from the society's point of view, i.e. they cannot distinguish between desired risks - as in "risk-taking"- vs. undesired risks - as in "risk-shifting". Delinquencies relative to loan-loss reserves did not evolve differently for banks with high and low MBS holdings. If anything, they have been slightly lower for banks with high MBS holdings. This suggests that QE did not lead to an additional buildup of financial instability.

Moreover, the results suggest that QE1 induced less tightening during a phase in which banks were generally tightening their standards. The findings suggest that banks increased risk in QE1 because of improved capital positions which should provide little concern with respect for financial stability. QE3, in contrast, induced additional easing of standards when the financial sector had been in a prolonged phase of recovery unrelated to improved capital positions. The additional risk-taking around QE3 could therefore provide somewhat more reason for concern about financial stability.