The Fed Is Opaque Because It's Flying Blind
On October 19, 1993, Alan Greenspan was doing what he did best, obfuscating. Called to testify before the House Banking Committee, Greenspan was questioned by Representative Toby Roth (R-WI) about rumors of verbatim transcripts of FOMC meetings. As would become his trademark later in the 1990's, Greenspan said a lot without saying anything.
Most of the exchange relied on the old politician's tactic of semantics. In the course of being queried about these transcript rumors, Greenspan repeatedly appealed to the assertion that "mechanical transcripts" simply did not exist. It was Federal Reserve policy, owing more to the primitive state of IT at the time, to record FOMC policy meetings and then re-record using the same tapes. The implication was obvious, that no recordings remained for historical FOMC meetings.
Thus the intentional focus on "mechanical transcripts" mystified the issue about records of what was said and by whom. Previous Fed regimes had adhered to the five-year tradition - where full and unedited written transcripts would be released after a five-year waiting period. That delayed transparency ended in 1976 under Arthur Burns, who had good reason to want to hide internal policy discussions.
Anna Schwartz, the famed economist, also testified the same day on the issue of transcripts, charging,
"The Fed has been lying about the transcripts for years. I could not discern whether [Greenspan] was talking about old transcripts or the regular meeting `notes' prepared after the FOMC meets. As far as I could tell sitting in the room, Greenspan did not clearly and explicitly disclose the existence of written verbatim transcripts so that members of the committee understood what he said."
The impetus for such drama was the smoldering debate about reforming the Federal Reserve System, with particular emphasis on policy and policymaking. The timing of the reform "movement" coincided with significant changes in the operations of the FOMC policy levers and a fuller interpretation of the still ongoing S&L cleanup. By 1993, the FOMC had run a course of then-record low interest rates, held artificially there in hopes of "stimulating" out of the then-unique occurrence of a "jobless recovery".
The same perceptions of economic malaise that contributed to George H.W. Bush's electoral defeat were now being pressed into Greenspan's domain. Part of that was due not only to the emphasis on "stimulus", but the means to achieve it. Sometime during the 1980's, the Federal Reserve changed its means of operation from targeting reserve or "money supply" levels to the federal funds rate; but nobody (even today) knows for sure when that happened or the specific rationale.
On January 26, 1993, Senator Byron Dorgan (D-ND), with co-sponsors Senators Kent Conrad (D-ND) and Harry Reid (D-NV), introduced the Federal Reserve Reform Act of 1993. Among other proposed changes, the bill intended to amend the Federal Reserve Act to mandate the FOMC's consultation with the Treasury Department, the Office of Management and Budget, and the Council of Economic Advisors on the preparation of reports. More importantly, it also sought to require immediate public disclosure of each and every change in intermediate monetary policy targets.
The bill itself made it to committee, but went no further. In all, it seemed an attempt to send a warning to Greenspan that Congress, Democrats in particular, were not so warm to the FOMC's secrecy and, by extension, the monetary emphasis on financial parameters. More than anything, there was a desire for answers to the economic questions lingering after the campaign. The Democrats successfully gained the White House but lost a net of nine seats in the House (though still retaining a large majority), and they wanted an account for the "jobless recovery" (or at least to tie it to the Republican embrace of monetarism in the course of less-favorable comparisons to other government interventions they preferred).
At the December 1990 FOMC meeting, right in the middle of the recession, the FOMC discussed the issue of targeting a lower federal funds rate, an official target, rather than changing quantities of borrowing activity through the New York Open Market Desk. By then, the mainstream thought inside the Fed was that changes in borrowing activity during maintenance periods were equivalent in every way to changing federal funds rates. But all through the latter part of the 1980's, particularly under Greenspan's Chair, there was worry about how markets, particularly wholesale money markets, would react.
Fed Governor Wayne Angell advised continuing the disingenuous stance of monetary directives, "as long as we maintain the charade of a borrowing targeting then that separation of function can still be there, it seems to me. That is, as long as we write the FOMC minutes based upon the charade of borrowing then the old practice can still be there." Rather than spook markets with undue concern over mechanical policy shifts, they would engage in willful deceit through altering the official "minutes" (while preserving the transcript with the understanding it would never get officially acknowledged or seen; most of the FOMC Governors and branch Presidents were unaware of the written transcripts).
To some, there is little distinction here as to what and why the Federal Reserve was so reluctant to publicly avow monetary changes. Fed officials had come to understand, in the context of monetary targeting, that there was equivalence to levels of open market activity and changes in the funds rates. But why was that such a problem?
In July 1987, it was noted that movements in the federal funds rate had become the de facto standard, at least in an informal manner, on the logistics of carrying out monetary policy. Governor Manuel Johnson noted,
"Many times the funds rate is used as a check to gauge whether the reserve estimates are correct. If the funds rate starts to move off from where we would have expected, given the reserve estimates, we seem to second guess our Treasury balance numbers and our excess reserve numbers and naturally assume that our reserve estimates are off because the funds rate is strange. Therefore, the Desk tends to do more or less in the open market."
There is more to it than that, however, as operational procedure was not directly translatable as policy. The Federal Reserve had turned to targeting money growth directly in an attempt to wrestle control of inflationary pressures in the late 1970's. But by 1982, the FOMC was increasingly concerned that the behavior of M1 was becoming fully uncorrelated with economic accounts, particularly GDP. In other words, their monetary theory was incomplete as the economy was behaving in a manner inconsistent with their definition of money, and its relationship with "demand".
The problem confronting policymakers in 1982 was high interest rates amidst a devastating double dip recession. Unable to better handle the relation of money and the economy, the FOMC had dual problems on its hands - to come to some kind of more "stimulative" stance in policy without sacrificing the inflation credibility they had accumulated post-1979 (earned or not). By outwardly and publicly targeting money supply, they hoped to continue the "charade" of credibility against inflation expectations in the markets and real economy. At the same time, by secretly moving to a target of borrowing or, in reality, interest rates, they could accomplish both goals as they saw it.
If inflation expectations continued to be tied to the perceptions about "money supply" growth, then all the better. As their data and research was showing, the basic concept of money supply was increasingly irrelevant to economic performance so it was an easy deception to maintain. But the FOMC recognized, perhaps for the first time in an official manner, that expectations were of primary importance. They could feed, so they believed, "stimulative" monetary policy through other direct channels, including wholesale money, without raising the specter of inflation through money growth. That made it a very valuable lie, worth maintaining in their cumulative estimation.
That wholesale money became the primary means of real monetary expression is the important point here. Even in the relatively early days of the 1980's, the Federal Reserve was beginning to see the changes in the banking system as they were occurring, without fully grasping all the implications. Money growth through traditional banking, M1 and M2 channels, were being replaced at the margins by the first rumbles of what would become the shadow system. And they wanted to embrace it, if only in secret at first.
By the time the S&L crisis and the recession of 1990-91 occurred, the traditional banking system was all but finished in terms of marginal growth. The mortgage market, the primary expression of domestic credit, was being transitioned fully from insolvent thrifts to the GSE's and their securitization conduits, while consumer credit was becoming a fast part of commercial banks and non-bank funding companies (relying on commercial paper).
Around the same time, the Fed under Greenspan was becoming more comfortable with the idea of these market changes and thus the emphasis on interest rate targeting (the market had largely figured it out by the late 1980's anyway). The borrowed reserves language remained in the FOMC minutes until 1991, during the Greenspan "stimulus" era of then-record low interest rates. It wasn't until February 1994, not long after Greenspan's questioning on the transcripts, that interest rate language was added, though it wouldn't be until 1997 that the explicit federal funds target was incorporated publicly.
Again, as much as this is a good story about Federal Reserve tendencies toward total secrecy, there is, I believe, something far more significant in these changes beyond that. In April 2011, the Federal Reserve published a paper titled, "Forecasting Recessions Using Stall Speeds". The idea here was to take GDP and GDI data to see if there was a limit or level where the economy upon entering a downward progression could not escape. There had been numerous instances of near-recessions scattered throughout economic chronology, so it seemed as if there might be some use in determining whether some kind of boundary applied.
Using GDI or GDP data without qualification, however, often led to "false positives", or periods where the economy slowed but did actually avoid recession. Curiously, there were clusters of these false positives associated with the "jobless recoveries" after the 1991 and 2001 recessions. These recovery indications were unique, and it forced the paper's author to warn,
"The dynamics at play in the early part of the 1990s and 2000s expansions may have been diﬀerent than the dynamics at play in the more-mature part of those and other expansions, and our stall speed models may have omitted an additional phase of the business cycle that has appeared in recent decades, namely the sluggish, jobless recovery phase."
Given what we know about the "recovery" post-2009, we can safely conclude another "jobless recovery" has been added to the cyclical routine in the domestic economic system. Further, the pace of the jobless recoveries continually wanes in progression; the 2001-03 recovery took longer and was less robust than the 1991-93 period, with the current phase far worse than either in every way.
The concurrent note to the observation of this apparent new economic phase (jobless recovery) was that recessions themselves have become more infrequent. It is suggested that this relates to monetary policy, and specifically the applications of "monetary easing" in the periods where false positives occurred during the jobless recoveries. In other words, there are hints here of monetary success in warding off what might have been recessions through new policy applications, especially the use of interest rates instead of other targets. That has led to elongated periods between recessions, including what passes for the Great "Moderation".
But therein lies another danger that the author actually expresses (where conventional economics largely remains silent),
"If the Fed anticipates and averts more recessions going forward, the stall-speed models may not work as well, in the sense that they may show more false positive stall readings. However, if the Fed does become better at preventing the types of recessions we have seen in the past, future recessions may start in ways that are diﬃcult to anticipate based on current data and knowledge. In that case, the stall-speed models again may not work as well going forward, but the economy would not necessarily experience fewer recessions."
What is being suggested here is that the economy may not always follow the models (or targets, in a rerun of the 1982 observations). If we extend this thought backward, it may also make sense in that the attempts at forestalling looming recessions in the jobless recovery phase might possibly be a contributing factor in the appearance of the jobless recovery phase in the first place. Going by strictly coincidence here, as the author is in assigning positive contributions in the diminished frequency of recession events, it could just as well be the case that monetary intervention has reached a point, including through mechanical changes in operations, where the normal/historical recovery process is itself interrupted.
The idea of stimulus-driven stability conforms to the conventional consensus that monetary policy has dampened economic volatility across the spectrum; less recessions at shallower depth, but slower growth, including less robust recovery. That interpretation appears to work except for 2008 and after (less volatility does not describe the panic period at all). Therefore, that highly suggests the monetary explanation is, at best, incomplete. Since this narrative cannot account for the depth of the Great Recession and the lack of any perceptible real recovery (outside of asset inflation), my interpretation becomes far more plausible in comparison (in addition to being intuitive).
The primary factor tying all these ends together is simply debt and credit production. I don't think there is any real mystery here as to the appearance of the jobless recovery in the age of central bank management of interest rates, nor do I think it simple coincidence. In incorporating monetary policy into the wholesale system, the FOMC set the conditions for massive "money" growth through non-traditional financial channels. The reason M1 was not as relevant in economic terms was that "money" growth happened increasingly in places like eurodollars and repos; shadow channels.
Interest rate targeting was a huge boost to those non-traditional channels, making the explosion in debt after 1990 almost inevitable. The appeal to a debt-based "stimulus" in recovery is simply not as effective as an organic recovery driven by actual price discovery and market signals through actual business. Further, what seemed to be a smoothing of the economic cycles, which Greenspan got full and unearned credit for, so much so that his intentional deviousness became "cute", was nothing more than artificial growth through debt. The huge decline in 2008 was belated recognition of that, and the lack of recovery after 2008 is nothing more than the structural shift as those previous debt channels have been obliterated.
Yet, there persists the monetary impulse to appeal to it again. Part of that relates to the other conclusions presented in the "stall speed" paper; namely that GDP or GDI signals are significantly augmented by using housing starts and the yield curve. In other words, an inverted yield curve, declining growth in housing, plus a downward level of either economic measure is consistent, without fail, with a developing recession.
If you are Chairman Bernanke in 2012 (or 2010) you might appeal to a steep yield curve and rising housing starts in the face of declining economic accounts to accomplish the same feat as his predecessor - not allowing re-recession to occur in the jobless recovery phase. It's not about the real economy as it actually exists, it's an attempt at recreating the academic understanding of how it all might work under these modeled assumptions.
What was revealed in 1982 is equally applicable thirty-plus years later. There is often a chasm between the academic understanding of economic affairs and the real economy in all its dynamic glory and mess. You might even have some sympathy as to why Greenspan would want to keep it all a secret, except that such secrecy also cuts in the other direction. It keeps the Fed from being exposed as flying blind as it creates new economic phases that did not previously exist out of monetary experimentations along the way. Or, in the current case, stretching the limit of just how "jobless" a recovery can be and still be consistent with such a notion.