A new study from the Bank of England addresses the question: does quantitative easing actually increase bank lending? And, if so, by how much?
Those are important questions, but one of the many lines of thought to which they lead is this: are some banks actually too large to bother rescuing? Maybe the policy makers who worried about the too-big-to-fail thing had it all wrong. The smaller banks help them do what they (wise policy makers that they are) presumably want to do. The bigger banks are irrelevant.
The monetary levers don't work on really big rocks.
Let's take this by steps. I submit we'll reach a conclusion that will interest, for example, those alpha hunters looking at long or short positions in the equities of banks.
One of the regular arguments in favor of the policy of expanding the money supply by central bank purchase of bonds, in the U.K. as in the U.S. and elsewhere has been that we (the wise policy makers) don't want to see banks just sitting on their cash. We want to see money where it can do some good. So let's give the banks a lot more of it, through this magic of central-banker fiat, and they'll start lending some of it out.
The Bank and Non-Bank Channels
The significance of the size of bank reserves and deposits as channels for the influence of QE upon macro-economic factors varies depending on how the easing policy itself is structured. The Bank of England certainly understood this when it was deliberating on the subject in March 2009. The new paper quotes the minutes of a monetary policy committee meeting at that time, where members expressed concern that buying gilts from banks would not help much, precisely because the pressures on banks given the "current strains in the financial system" meant that banks were "less likely to increase their lending following an increase in their reserves."
That is why the committee decided that asset purchases would do more good if they were made "from the domestic non-bank financial sector rather than from banks."
That month, the B of E reduced the bank rate to the historically low level of 0.5% and introduced a gilt purchases policy that would continue (in its first round) through January 2010. That first round entailed the purchase of £200 billion of assets, roughly 14% of the country's nominal GDP.
In a second round, beginning in October 2011 and continuing through November 2012, the Bank purchased an additional £175 billion of assets.
As the author of the new study observe, though, there can't be any very firm dichotomy between the bank channel and the non-bank channel for QE. After all, if non-bank financial institutions sell their gilts to the central bank, the size of their deposits in banks will increase, the size of the bank's liquid holdings will increase, and this will make banks "less reliant on seeking other funding to manage their liquidity needs."
Back to the issue of size
The study treats in particular the ratio of a bank's deposits over its total assets (the DA ratio), which they take as a good proxy for the effect of QE. It reaches the following conclusions: first, that there was a statistically significant influence in bank lending as a consequence of the money creation; second, that the effect was limited. In part, the effects were small because the marginal effects through deposits are small. In part, too, the effect was small because there was something less than a full pass-through from QE to deposits.
Further, at a micro level the effect depends on the size of the bank involved. The "major banks account for the dominant share of total lending in the U.K. banking system," the report says, so this heterogeneity is important. If one isolates the large banks, the positive effect of QE on lending ceases to be statistically significant. So in effect all the increase in lending created through QE actually comes through the small banks.
Why? Well, probably as the study says because "small banks have limited access to capital markets to raise non-deposit liabilities and therefore a given negative shock to their deposits leads to a sharper reduction in lending than is the case for large banks."
This intrigues me. If the world's ruling elites have decided that lending needs a boost, and they are figuring out that big banks are irrelevant to the mechanism by which they can boost it, then shouldn't this contribute in turn to the notion that the bigger banks are irrelevant enough to be allowed to fail?
Perhaps the gravy train of the 2B2F implicit backstop is finally chugging through its final stations.