In my last post regarding Ben Bernanke's memoir, I took Bernanke's Fed to task for electing to sterilize its pre-AIG emergency lending, thereby making sure that, while it was rescuing a small number of troubled firms, it was also reducing the liquid reserves available to others. It was doing this, moreover, at a time when increasingly worrisome economic conditions were giving rise to exceptional liquidity demands. The predictable result was an overall shortage of liquidity, which manifested itself in a collapse of bank lending and spending.
I now turn to an even more mind-bogglingly wrongheaded step taken by Bernanke's Fed in the course of the financial crisis: its decision to pay interest on banks' excess reserves.
The Fed began paying interest on reserves (IOR) in mid-October 2008, just after ending its program of sterilized lending. That these steps coincided was no accident, for interest on reserves was meant to be a substitute for sterilization, aimed at the same result, namely: that of making sure that the Fed's gross asset purchases did not give rise to any corresponding increase in bank lending.
The Fed resorted to IOR because, by the time of Lehman's failure, it had reduced its Treasury holdings to their practical minimum. Consequently, when it came to bailing-out AIG, the Fed found itself in a quandary: the bailout would mean an increase in the overall size of the Fed's balance sheet, and a like increase in the monetary base. Other things equal, the increase would mean a loosening of monetary policy. Yet the Fed was determined to avoid such a loosening.
It was to escape from this quandary that the Fed came up with the oh-so-clever idea of rewarding banks for not lending their otherwise unneeded reserves in the middle of one of the twentieth-century's most severe economic contractions. You needn't take my word for it. Here it is from the horse's mouth:
We had initially asked to pay interest [in 2006] on reserves for technical reasons. But in 2008, we needed the authority to solve an increasingly serious problem: the risk that our emergency lending, which had the side effect of increasing bank reserves, would lead short-term interest rates to fall below our federal funds target and thereby cause us to lose control of monetary policy. When banks have lots of reserves, they have less need to borrow from each other, which pushes down the interest rate on that borrowing — the federal funds rate.
Until this point we had been selling Treasury securities we owned to offset the effect of our [emergency] lending on reserves (the process called sterilization). But as our lending increased, that stopgap response would at some point no longer be possible because we would run out of Treasuries to sell. At that point, without legislative action, we would be forced to either limit the size of our interventions…or lose the ability to control the federal funds rate, the main instrument of monetary policy…[By] setting the interest rate we paid on reserves high enough, we could prevent the federal funds rate from falling too low, no matter how much [emergency] lending we did (Courage to Act, pp. 325-6).
Nor does Bernanke's understanding of the Fed's action differ from that of other Federal Reserve authorities. In December 2009, for example, Richmond Fed economists John R. Walter and Renee Courtois offered an almost identical account. The Fed's emergency credit injections, they wrote,
had the potential to push the fed funds rate below its target, increasing the overall supply
of credit to the economy beyond a level consistent with the Fed’s macroeconomic policy goals, particularly concerning price stability.
For a while sterilization solved the problem. But
following the failure of Lehman Brothers and the rescue of American International Group in September 2008, credit market dislocations intensified and lending through the Fed’s new lending facilities ballooned. The Fed no longer held enough Treasury securities to
sterilize the lending.
This led the Fed to request authority to accelerate implementation of the IOR policy that had been approved in 2006. Once banks began earning interest on the excess reserves they held, they would be more willing to hold on to excess reserves instead of attempting to purge them from their balance sheets via loans made in the fed funds market, which would drive the fed funds rate below the Fed’s target for that rate.
There you have it. The Fed paid banks to hold excess reserves, so that they would quit lending them, in order to make sure that the "overall supply of credit" did not exceed levels "consistent with the Fed’s macroeconomic policy."
And what level of credit supply was it that the Fed was so anxious to not "exceed"? Perhaps the following chart will give you some idea:
If you are starting to forgive me for using the term "wrongheaded," I have made my point.
When the Fed first started paying interest on reserves, it did so at a rate exceeding the rate at which banks were prepared to borrow from one another in the overnight ("federal funds") market. Because this was the case, banks more-or-less withdrew from that market, leaving only some GSE's, which had reserve accounts at the Fed but were not eligible for IOR, to participate in it. The observed average overnight rate for transactions among these institutions — the so-called "effective" federal funds rate — has been consistently below the interest rate on excess reserves:
Needless to say, after most banks withdrew from the overnight market, the overall volume of overnight loans declined substantially. Remember all that talk about how Lehman's failure led to heightened concerns about counterparty risk, which caused the interbank market to seize-up? Well, that's not what happened. Although some large banks had to cut-back on overnight borrowing in response to Lehman's failure, small banks actually borrowed more. A Liberty Street Economics post, from which the following image is taken, supplies details.
It's possible, of course, that the banks that were flush with excess reserves wouldn't have lent those reserves even if they didn't bear interest, because those banks were also short of capital. The thesis is at least plausible, since the Fed supplied fresh reserves to capital-starved firms by swapping them for mortgage debt and other illiquid assets. The exchange reduced the cash recipients' (risk-adjusted) capital requirements, but did so only if they held on to the cash. As an alternative to having to raise new capital, the swaps were a bargain — and especially so once reserves bore a modest return.
According to Huberto M. Ennis and Alexander L. Wolman, capital shortages did at first discourage banks that held excess reserves from lending them. But already by late 2009 some "would have been able to use reserves to accommodate a significant increase in loan demand without facing binding capital constraints." Two years later, the same authors report, "a significant proportion of the reserves held by large banks…could have been quickly used to fund loans without pushing these banks against their minimum regulatory capital levels."
The reserves "could have been" used. But they weren't. Why not? Ennis and Wolman conclude that, for banks that were no longer capital constrained, limited "loan demand was likely the main driving force behind banks’ lending behavior." But that is just another way of saying that the anticipated return on loans was low compared to the return from not lending — that is, compared to the return on holding reserves.
It remains possible nonetheless that IOR made little difference, because the return on loans would have been below the return on reserves even if the latter bore no interest. That at least some "natural" interest rates, including the natural overnight rate, became negative during the crisis, and that a few may still be negative today, is the belief of more than a few economists. That includes authorities in charge of several European central banks. What's more, it includes Janet Yellen, who in a recent paper observes that "the natural real rate fell sharply with the onset of the crisis and has recovered only partially," and supplies the following chart summarizing extant estimates of the natural ffr:
If these estimates, or at least some of them, are correct, banks would have withdrawn from the federal funds market by early 2009 even without IOR.
But several caveats are in order. First, note that the natural ffr estimates turn decisively negative only in late 2008. This suggests that, if, instead of introducing IOR in the first place, the Fed had allowed its post-Lehman asset purchases to translate into increased bank lending, and especially if it had not sterilized its asset purchases before then, the natural funds rate might never have gone negative. As I pointed out in my post on sterilization, a collapse in aggregate demand means, among other things, a collapse in the nominal demand for all sorts of credit, and a corresponding decline in market-clearing nominal interest rates.
Second, and no less importantly, while overnight lending rates may have turned negative, not all lending rates did so. Returning the interest rate on reserves to zero would therefore have encouraged bank lending, even if it failed to encourage overnight lending. The one qualifier is that, because the supply of overnight funds itself would have remained constrained, banks would still have had a greater than usual need for excess reserves to meet their settlement needs.
Finally, even allowing that nothing short of negative interest on excess reserves would have inspired banks to lend those reserves, it remains the case that IOR was a bad policy, in the sense that abandoning it would at least have been a step in the right direction.
For my part, I do not doubt that, whether negative IOR would have been ideal or not, positive IOR played an essential part in the vast post-2008 accumulation of bank excess reserves — an accumulation that proceeded in lock-step with the Fed's large-scale asset purchases, thereby allowing the Fed to add trillions to its balance sheet, without contributing a jot to bank lending:
Here is where I must part company with some of my Market Monetarist friends. For while those friends hold, as I do, that IOR was counterproductive, and that Fed asset purchases might have been far more effective in boosting recovery without it, they have tended nonetheless to defend the Fed's large scale asset purchases ("quantitative easing"). I think this was a mistake, both because it meant accepting the Fed's own dubious (and hardly "monetarist") theories about how LSAPs were supposed to aid recovery despite the non-lending of added reserves, and because it overlooked the very real adverse effects of those purchases, including the sacrifice of ordinary means for monetary control that they entailed.
There are, I realize, some who argue that IOR isn't really equivalent to paying banks not to lend, and that it is therefore not to blame for their having accumulated so many excess reserves. Their arguments are, if anything, even more screwy than the logic — which those arguments manifestly contradict — underlying the Fed's decision to resort to IOR in the first place. But this post is long enough, so I'll turn to these arguments in Part II.
As it happens, just as I am completing this post, Janet Yellen is holding a press conference announcing what she regards as the Fed's first steps toward the "normalization" of monetary policy that has become a policy desideratum in the wake of the Fed's highly abnormal marriage of IOR and large-scale asset purchases. Of what do these steps consist? First, the Fed will raise its federal funds rate target by a quarter of a percentage point — a meaningless gesture, given (as Yellen herself understands) that the target was already above the "natural" funds rate before the hike.
Second, the Fed plans to double the rate of interest on excess reserves.