Official Denial and Happy Tones Are Never More Ridiculous Than Now
LIBOR is not a popular topic for monetary policymakers. To begin with, it refers to eurodollar funding not dollars. Since orthodox canon denies any particular relevance to offshore markets the Federal Reserve would prefer to keep to its “don’t ask, don’t tell” type of policy for them. There are occasions, however, when doing so starts to get people wondering about all the “wrong” things.
Before getting to the latest bit of brazen unreality, first we have to address our frame of reference. We are taught maturity transformation from the very first day of Economics 101, or whichever course gets to the money multiplier idea. A bank can take advantage of the differences along the yield curve. By borrowing very short-term, the cost of funding is much lower than the interest received by lending those funds for a longer period of time.
Interest rate targeting therefore makes some intuitive sense. A central bank that can control the cost of funding can exert influence over financial factors starting with lending. If the Federal Reserve, for example, may raise the federal funds rate, the theory tells us that banks will curtail lending for the increased cost, and lowered profit, of maturity transformation. By doing only that, monetary policy is meant to effectively, and predictably, control downstream outcomes – including marginal changes economywide.
But interest rate targeting is already a plug-line or shortcut. The theory sounds nice but it is nothing more than backwards engineered from reality. Economists using regressions noticed correlations between the movement of federal funds and economic aggregates. Whether or not those latter were directly related to alterations in the monetary policy target has never been established. It has always just been assumed so and really on the basis of “what else would it have been?”
It’s not quite an original sin, but a sin nonetheless. Monetary evolution forced central banks to abandon their strict definitions for money. Because of that, what else would monetary policy use to accomplish its mandates? The more scientifically valid answer would have been for authorities to completely and thoroughly investigate that evolution. A shortcut was just easier.
This policy regime was something altogether different. It shifted emphasis from monetary inputs to economic outcomes. Economists believed, essentially, they wouldn’t have to define or measure money, or even gain substantial knowledge about how it all worked in practice, so long as they could correlate their interest rate target changes to economic aggregates like GDP growth, inflation, and the unemployment rate. This is Positive Economics, or what today is called econometrics.
This used to be, and still is, considered sound practice. But it’s a bit like hiring the cheapest firm you can find to get your rocket ready to fly into outer space; you don’t care what they put into the rocket or how they arrange the internals, you only care that when the ignition button is pushed it stays within your margins for flight control, as you only expect that it will. If it starts to stray, well.
According to the very basic premise of maturity transformation, a higher money rate is tighter than a lower one. The Fed raises and lowers the federal funds rate, dictating which is which by the direction.
That’s not really how it works, obviously.
On August 8, 2007, 3-month LIBOR was 5.38% (a weighted average of surveyed participants in the LIBOR panel among the British Bankers Association). On September 29, 2008, 3-month LIBOR was 3.8825%. In between, the Federal Reserve had brought its interest rate target for federal funds down from 5.25% to 2%. We know by what happened that nominal rates may seem at times to tell us almost nothing.
For one thing, on August 8 the world appeared to be somewhat normal. In sharp contrast, on September 29 a worldwide panic had begun; the Dow fell 777.68 points, the largest single-day point drop in its history (until recently); the MSCI World Stock Index gave up 6%, its worst day; London’s FTSE fell 15%; while other stock markets around the world experienced similar outcomes (Brazil’s, for instance, was halted after a 10% drubbing).
Things, near to the point of everything, were being liquidated and it would not stop for several more weeks.
Confusing matters even more, the federal funds effective rate (EFF) was doing something very strange. This latter interest rate is the weighted average yield of actual transactions that take place in NYC federal funds, an average that is supposed to be close to the monetary policy target. If it strays the implicit threat of central bank action (open market operations) becomes explicit action.
This, too, is what we are taught in Economics 101. To coordinate the target and the market the Fed will buy and sell assets in its portfolio, raising and lowering the level of bank reserves by doing so, such that the effective rate is normalized as an average to the policy target. Up until August 2007, there was never a need for actual transactions; the Fed stated its policy and the money markets conformed on their own to it.
But as had occurred in 2007, EFF began to once more exhibit a terrible tendency toward shallowness. A monetary policy target has to work in both directions. This is not optional. That means if the target is 2%, you don’t want the rate to be too far above or below. It has to stay close on either side.
Not so in later September 2008 (as well as the months following). Effective federal funds had been 1.56% on September 29, 2008, even though the target was 2%. The Friday before that particular Monday, September 26, EFF was 1.08% with the same target.
Therefore, what really matters to us as a matter of interpretation of real world conditions is the spread, not the nominal money rate in isolation. A negative spread of -44 bps in federal funds on the same day as a huge positive spread of +188 bps in 3-month LIBOR tells us quite a lot that the overall reduction in nominal rates all across those parts does not. There were “too many” dollars in NYC and a catastrophic shortage of them offshore, though Wall Street would sorely disagree with the first part of that characterization even as that was the effect.
But if what matters is the spread, we better have some good understanding of it even beyond that second-tier analysis. We are taught in Economics 201, or whichever, that a positive spread like that means credit risk. In LIBOR, a big plus sign is supposed to convey the view from market participants of each other’s creditworthiness. In the conventional narrative, they pulled back from unsecured lending (and secured lending, too, for that matter) because they were afraid of what subprime “toxic waste” the others might have hidden somewhere on the balance sheet.
This was, however, a false conventional impression. Credit risk was never the primary consideration as the whole subprime thing got way more attention than it ever truly deserved. This is not to say that it was some minor nuisance; it wasn’t. But what led to panic had less and less to do with subprime and credit risk and more and more to do with global monetary practices and the kind of money structures then in place.
Here’s a real-world example of what I mean. In its regulatory filing (10-Q) for the third quarter of 2008, Citigroup reported net interest revenues for the first nine months of that year of $40.4 billion. That was up sharply from $33.1 billion reported in the first nine months of 2007. This rather substantial gain for the firm’s traditional bank business was that maturity transformation spelled out in Economics 101: nominal yields had declined far more than those for the loan and security assets sitting on the asset side of Citi’s balance sheet. Well done Bernanke?
If banks had been nothing more than actual banks, end of story with its happy ending. But Citigroup, as all the rest, has another side which depends on prices not rates (except rates that input into prices). Non-interest revenue was $6.8 billion in those nine months of 2008, compared to $38.9 billion during the same months of 2007. Monetary policy had delivered +$7 billion directly in so-called stimulus, but that was completely overwhelmed and obliterated by the shadows. In less than two months after reporting these figures, Citi was handed the largest bailout in American history.
For comparison with money dealing shadow firms that had no depository banking to begin with, such as AIG, there was no channel for monetary policy at all. AIG’s total revenue in the first half of 2008 was $34.0 billion. In the first half of 2007, the firm reported $61.8 billion. How many companies of any kind can survive on a 50% revenue reduction? You can claim as many did contemporarily that AIG wasn’t the Fed’s job, but LIBOR and what was really behind it especially in comparison to EFF left them no choice.
Reality wasn’t like the textbooks.
Citi’s $32 billion reduction in non-interest revenue and AIG’s topline collapse had little and often nothing to do with subprime, and by the end it had even less to do with cash losses. These were almost always OTTI impairments, accounting issues about pricing, hedging, and control of asset positions being dictated by increasingly illiquid markets and market prices. On the outside, it appeared as if subprime was creating uncontrolled massive losses; on the inside, it was the revocation of balance sheet capacity as liquidity and money locking the system into the same self-reinforcing downward spiral typical of every bank run in history.
Money markets were never truly a monolithic whole. These were always fragmented if operating prior on an apparently seamless basis. If there had ever been a surplus of funds in federal funds as compared to eurodollars and LIBOR, money dealing banks on either side of the Atlantic (and often the same bank going between headquarters and its foreign subsidiary) would have arbitraged any spread in either direction. This is what’s called covered interest parity.
To undertake arbitrage of any kind requires balance sheet capacity (and I’m being purposefully vague about it here for space constraints). In other words, the same thing ruining AIG and Citi’s revenues in 2008 was reducing the system’s ability to operate according to covered interest parity (we also saw it in places like interest rate swaps, with negative swap spreads). What everyone including the Fed and the Economics 101 and 201 textbooks assumed as a single global money market had been pulled further and further apart into its constituent pieces, each of which fared differently according to its own constraints (including and especially geography).
Thus, LIBOR’s positive spread, and EFF’s negative spread for that matter, related to liquidity not credit risk; “dollars” and balance sheet capacity, not subprime and “toxic waste.”
One of those parallel spreads that policymakers came to rely on (unnecessarily, in my view) was LIBOR-OIS. Here’s the FOMC describing it on an emergency conference call held September 29, 2008:
“First, we’ve seen a sharp rise in overnight dollar funding rates and in term LIBOR–OIS spreads. For example, on Friday, the three-month LIBOR–OIS spread was over 200 basis points, and in fact, LIBOR may actually understate the degree of funding pressure.”
As I wrote at the outset, policymakers avoid talking about LIBOR until they can’t.
One of those times is now; or, at least at the last policy meeting in March 2018. The minutes of that gathering were released this week, and tucked away within them was this gem:
“In short-term funding markets, increased issuance of Treasury bills lifted Treasury bill yields above comparable-maturity OIS rates for the first time in almost a decade. The rise in bill yields was a factor that pushed up money market rates and widened the spreads of certificates of deposit and term London interbank offered rates relative to OIS rates.”
LIBOR-OIS has blown wider of late, understandably upsetting a good many people because of the last time it had happened. The Fed’s discussion gives us a convenient and benign excuse for why we should ignore it this time (note the weasel term “was a factor”).
According to their view, tax reform and the government’s already massive deficit requires a substantial increase in the auction of Treasury bills (bills are used first for any changes in the government’s debt financing needs). A heavier supply of bills, ceteris paribus, should increase rates on bills. And since these are treated often as money equivalents, an interbank participant would rather place short-term funds in higher yielding bills leaving other money market rates to rise in tandem.
Sounds reasonable enough. And it’s complete bull. Demonstrably so.
On January 30, 2018, the 4-week T-bill equivalent yield (like LIBOR and EFF, there is no single 4-week Treasury bill, the equivalent yield is a bootstrapped, meaning calculated, rate of what one would be if there was) spiked from 1.28% to 1.49%, a highly unusual move for any bill on any single day. What was it that happened the day before?
This last stock market liquidation began on January 29. As if to emphasize the nature of what took place, this all repeated again a week later. February 5, 2018:
“The Dow lost a staggering 1,175.21 points on Monday [Feb 5] to close below 25,000, giving up all its gains for 2018. Further, this also marks its biggest drop in a single session, in terms of points since its inception. Such a decline was also the blue-chip index’s biggest percentage drop in a single day since August 2011.”
The day after that? The 4-week bill yield spiked again, this time from 1.41%, about where it had stayed after the initial rise, to 1.48%. Liquidations in stocks and then liquidations in bills (aside: ask yourself who holds the largest pile of Treasury bills on the planet?)
Also on and after February 6: 3-month LIBOR accelerates, blowing out (further) the spread to OIS. If you still think of a higher spread as credit risk, then why not government deficits and money equivalents (though, even then the timing is still backward)? If you instead learned from 2008 that these spreads are all about liquidity and liquidity risk (and therefore, you know, liquidations) LIBOR-OIS seems perfectly consistent with what actually happened – and might still happen (downside risk).
The worst part of all this isn’t just the indications of another looming, escalating global “dollar” problem. It’s why there might be one.
One big reason is the Treasury bill fiction the FOMC chose to place in its official meeting minutes. They’ll point to “was a factor” if ever they have to, but there is no doubt this is the explanation they prefer to put out and for everyone to just accept. People are talking about LIBOR-OIS and so they had to say something. Anything. But they couldn’t declare it was exactly what it appears to be because that would run contrary to everything they are doing, saying, and quite often take on nothing more than faith.
Instead, they deliberately throw out this seemingly nondescript reference because it will serve as a de facto official explanation (without appearing to make a big deal out of it) while at the same time they know it will never be challenged. Even though it is, again, demonstrably false, the mainstream media and almost every form of financial commentary is deathly afraid of something like OIS; they don’t understand it, they believe nobody else does, so therefore if the Fed says something about it, then it must be so. Leave it to who are still treated as experts. Failure isn’t just an option, it’s become the job description.
Say the word “swap” and people go running for cover. OIS is the overnight index swap, and what it is really doesn’t even matter. Complexity has been a shield for this massive dereliction and for a very long time. It’s how you get an entire decade where conventional wisdom still thinks of 2008 as a subprime mortgage problem. It’s how the entire global economy could have shrunk, and then stayed that way through intermittent but serious further monetary problems no matter how much QE undertaken wherever.
We’ve been through two more of these (2011-12 and 2014-16) since the original crisis. All the signs are aligning in the direction of a third. These things don’t happen all at once, taking time to develop (even the panic was a rather lengthy process). Official denial and happy tones are never stronger, or more ridiculous and brazen, than in these times.
