When decrying the state of monetary policy that relies on essentially a “dead” money market, what does that actually mean? The FOMC, after all, is using the federal funds rate to “tighten”, ostensibly, even though there isn’t anybody there. They have developed other tools to go along with the federal funds rate, but all that does is highlight the central problem. This should have been far more appreciated as a central factor in almost everything that has occurred, financially, since 2007.

When discussing money markets everyone writes and speaks the “s” on the end but conceptually visualizes a singular mass. There are lots of different money markets, but almost always they are assumed to be an interconnected, operable whole. Thus, if Janet Yellen decrees that the federal funds rate is x, it assumed and expected the rest is just carried out however that might be.

Except that the panic in 2008 showed us the folly of such thinking. In general liquidity terms, starting August 2007, what was believed to be a monolithic whole, “the” money markets, fragmented and then blasted apart into each constituent piece. The very first indication of systemic contagion on August 9, 2007, was exactly that point – fragmentation via geography, or that there was a very real distinction between domestic and offshore “dollars.” It would continue, as repo fails showed that collateral was just as much a money market as either federal funds or eurodollars (and then credit default swaps and dark leverage betrayal demonstrated just how far it had all evolved).

It should not have caught so many by surprise, but it was at least understandable given that the operation of various money markets and integration among them had been largely smooth and uninterrupted for decades. Still, history of how money markets developed in this modern sense indicated all along the potential for this nearly fatal flaw (history is sadly discarded and left unappreciated largely because successive generations view their own contributions as dispositive solutions; instead they maintain all the premises of the ages-old problems and only give them new terminology and visit new ways to express the same issues).

The repo market grew up in the 1950’s as the federal funds market roared back to life, having remained dormant throughout the Great Depression, World War II and the immediate post-war years. It was truly a simple after-effect of regulation owing to how the early 1930’s banking collapse was to be dealt with. Non-banks were prevented from accessing federal funds as a matter of practical separation – keeping dealers and commercial banks separate from depositors. Dealers, however, had to access funds somehow and since the call money market that had functioned as the backbone of securities funding for almost a century had been effectively banished after the 1929-33 collapse, these institutions turned increasingly to repos.

The technical distinction of a repurchase agreement (repo) has, too, been lost to history as at one point it was actually a purchase and a sale – it was only reclassified as a loan, collateralized, in September 1956 by decree of the Comptroller of the Currency (instructing his bank examiners, also capturing state-registrations, to treat repurchases as a loan as they were intended). Dealers had been using repos of t-bills (since government debt had come to dominate the securities markets after WWII) to tap non-bank sources of funds as well as funding outside of NYC (which still had seasonal and structural points of illiquidity).

With the rise of federal funds volume (still mostly connected to NYC) in the 1950’s, larger banks began to address the fragmentation and the opportunity given by a more fluid and liquid market for bank “reserves.” Because non-banks (a technical and legal distinction) were prohibited from directly participating and because banks were restricted as to the interest rates they could offer non-bank deposit accounts, repos became a popular tool to increasingly tap non-bank and non-NYC funding sources while giving non-banks an alternative to generating interest exclusively in other formats. In other words, banks engaged non-banks (corporations, municipal governments, securities pools) in repos at an interest rate slightly below federal funds so that non-banks could earn short-term something like money market participation (instead of strictly t-bills).

From the bank’s perspective, the repo was free of reserve requirements and interest controls so it was mutually beneficial as a source of bank funding. Over time, the repo rate (under most circumstances) began to trade in close proximity to the federal funds rate for these reasons. Because of that, the idea of these fragmented pieces of money markets becoming a singular whole was likewise turned to conventional wisdom and treated as such (particularly by monetary policy shifting to interest rate targeting, of the federal funds rate, in the 1980’s).

However, as the history of these markets amply demonstrates, they require action, activity and resources of intermediation to become apparently so. The very word that should most be applied to banking, but no longer is, is “intermediation” – literally meaning standing between two or more factors and bridging the gap. Money markets are not something themselves, they are the outcome of action and financial production.

In money markets, liquidity and flow depended upon first banks turning non-banks into federal funds via repos in order to navigate regulations and a great many disparate incentives. That need has never changed even though the regulations and the institutions making the effort have; and money markets have only expanded in size, reach and further distinctions. The great crash in 2008 was due to the fact that money markets finally showed their distinctions – all because money dealers were no longer willing to dedicate the resources and take the risks of turning often-contradictory markets into a seemingly flawless and seamless whole.

Long after it was too late, central banks stepped in to try to reconstitute money markets once more as a unified mass of common function; it didn’t work, so central banks simply became money markets (in large part). The problem of that should be more obvious especially in the context of current monetary affairs – central bank intermediation in money markets is nothing like money dealing intermediation of private money markets. Central banks are bureaucracies and act like them, a rigidity that isn’t well-suited to the dynamism required to bridge so many divides and make them work together.

So the first moves in this FOMC “tightening” have left some confusion as to why such “tightening” hasn’t done much to liquidity. Via ZeroHedge:

Recall that two weeks ago we cited repo-market expert E.D. Skyrm who calculated that moving general collateral higher by 25bps would require the Fed draining up to $800 billion in liquidity: “In 2013 on my website, I calculated that QE2 moved Repo rates, on average, 2.7 basis points for every $100B in QE. So, one very rough estimate moved GC 8 basis points and the other 2.7 basis points per hundred billion. In order to move GC 25 basis points higher, in a very rough estimate, the Fed needs to drain between $310B and $800B in liquidity.”

 

Then on Wednesday morning Citigroup opined that the liquidity drain could be substantially greater: “There will be a separate document from the NY Fed with details around the operational aspects of the liftoff. Of primary interest will be the size of the overnight reverse repo facility that the Fed will put in place to pull short rates higher. We don’t think it will be unlimited, but a size large enough that will keep short rates from falling below the 25bp floor – and the size could be as high as $1tn.”

 

So what happened?

As I noted last week, the Fed “threatened” as much as $2 trillion in its reverse repo, but there was barely an uptick as Friday’s RRP, two days after the policy change, was just $143 billion. So the Fed raised rates and only the “deadest” parts of the money markets obliged while a great money other parts have openly resisted (subscription required).

Again, going back to what I wrote in September 2013 when the RRP program expansion was announced:

The reverse repo facility is an attempt to address systemic weaknesses exposed five and six years ago, cloaked in the context of thinking about an exit from QE. Instead, it is a half of a liquidity bridge that still does not fully solve fragmentation, but at least the Fed will (theoretically) have more control over short-term money markets. That can’t be a bad thing, right?

Central banks simply cannot perform the “correct” intermediation in order to recreate money markets as they existed before August 2007; it is just impossible. That means either they have to find an alternate solution or get used to being the permanently lacking ad hoc workaround. Neither of those should be considered anything like “normalization” not least of which is because the Fed isn’t prepared for any of it. As far as Janet Yellen seems concerned, in public, money markets will just spring back to life as if they were there all along – exactly like her view on the economy despite all its revealed challenges in the past year or so.

This harried and backwards nature is actually standard procedure at the Fed. “We” only think that it has expert capacity and solid working knowledge of these kinds of things because that is the reputation it has purposefully and forcefully cultivated instead of courting useful and scientific knowledge. That was 2008; that was also eurodollars and even federal funds itself. Despite the fact that federal funds, newly “borrowed” reserves and repos had experienced huge growth throughout the 1950’s it wasn’t until May 1959 that the Fed decided to undertake a serious study with which to understand exactly what was going on. The study would last three years and it wasn’t until August 1964 that the Fed had developed and was ready to incorporate statistics into its policy framework.

Is it any wonder the Great Inflation would begin a year later? Central banks are bureaucracies and act like them. Money markets require something entirely different and increasingly it is looking as if they will not (never?) get them. They really don’t know what they are doing – they never have.