With close experience to the “dollar” run in July and August, we should expect that at least the contours of that disorder would be visible in the TIC data for July. The figures provided by the Treasury Dept. did not disappoint. In fact, the “dollar” waves themselves become almost fully visible in the data here and have really illuminated the pathology. If you start from the position of the “global dollar short” and realize that the rise and fall of the dollar relates to the relative expensiveness of maintaining that short across the globe, the pieces fit perfectly together (almost).
What TIC provides is really a netted figure for the regular churn of reported dollar-denominated assets of all kinds; not just UST. In January, for example, there were $2.324 trillion in reported purchases but $2.364 trillion in reported sales, equating to a net decline of $39.6 billion. That was a steep drop in foreign holdings, which in convention suggests foreign holders “selling” dollar assets. That may be true in its most technical aspect, but in terms of global eurodollar conditions it meant that the various financial and non-financial “dollar” channels were mobilizing their own sources for maintaining a steady (or at least orderly withdrawal) supply of short “dollar” liabilities.
Of the $39.6 billion decline in TIC flow that month, $27.6 billion was recorded as “private” and $12.0 billion as “official.” That makes some good sense given that January was a horrible month for the “dollar short”, leading to the SNB giving us our first major central bank “defeat” at the hands of the rampaging eurodollar. The fact that it was in TIC on both sides of the private and official divide describes the broad and uniform nature of duress.
Ever since January, net TIC flows have been much more positive suggesting the intermission between the “dollar waves.” The update for July, then, gives us the outlines for the second wave (this year) that showed up in LIBOR, repo and elsewhere. Total net flows, which had been as high as +$87.2 billion in June suddenly sunk to neutral in July; just +$4 billion.
Even that barely positive number belies the eroding subtext. For all the positive flows on the private side, central banks (“official” accounts) have been straining to provide “dollars” consistently throughout this year. That is especially true in UST assets, as there has been negative flow into official UST holdings in all but one month (May) going back to, and including, the turmoil in December. This disparity (that private flows could return positive while official flows did not) this year in flows suggests itself the underlying cause: global financial conditions.
Certainly the main contribution toward the turmoil in January was the record “outflow” of bank liabilities in December. I have been talking about quarter-end bottlenecks in “dollar” conditions and flow for some time now with good reason. The reported figures for bank “dollar” liabilities are far from comprehensive, but given what has been easily observed in that direction this year in real time (or close to it) the correlation is both astounding and confirming. The three main “events” in the “dollar” so far have been October 15, January 15 (along with the events in December, including the nadir of the junk bond selloff on December 16) and the last “dollar” run that began around July 6 inside the two weeks after the end of Q2.
What the TIC figures reveal is that the contraction in reported bank liabilities in the three quarter-end months just prior to those three major “dollar” events were massive, record declines.
The space in between, Q1 entering Q2 2015, wasn’t nearly as severe – about half the scale of that illiquidity trio. Our “dollar” bottlenecks appear as the degree to which window dressing overwhelms the positive liability creation from the entire prior three months.
I should clarify at this point that the end-of-quarter bottleneck is not a recent phenomenon; in fact, it has been a constant feature for decades, which is why these types of market events cluster around these timeframes. The difference now vs. any other quarter end is how structural or cyclical changes amplify and reveal these liquidity cracks to the point of the severity we have witnessed for the past year or two (really going back to 2013). When combining all three months into a comprehensive quarterly liquidity picture, these reported balances for bank liabilities reveal the “dollar” waves in obvious fashion (including the first in the middle of 2013).
Again, this reveals that the problem is almost strictly financial, as in bank balance sheet “supply.” What the official and private accounts of various dollar assets show is essentially how the world is dealing with the ups and downs in this financial decay. Since January, it has been central banks under constant pressure where the “private” side seemed almost undisturbed (until July) to the point of acting out normalcy.
While China has received note for its now-open UST sales, it has not been alone or unique in that regard. Indeed, there have been several surprise “partners” in central banks (or national governments, which is not always one and the same) contributing “reserves” in the face of financial retreat in global “dollar” supply. Straight away, the UST holdings by Japan show the grim reality in far closer proximity, in this respect, to China (including, as ZeroHedge has pointed out, their proxies in Belgium) than seems otherwise appreciated.
From Brazil to the Oil Exporters (the misinformed petrodollar), the “dollar’s” struggles this year are quite evident; which is why economists that have referred to this as a “strong dollar” are continually surprised by economic and market downsides.
The case of Brazil is especially poignant, including its economic relation to all this, as the TIC trend shown above is augmented by the reality of what its central bank tried to do financially. The real’s depreciation and devastation in 2015 has been an order of magnitude worse than that of 2013 at the start of the first “dollar” wave, but you wouldn’t know it from the TIC flows which look about equal or even slightly more severe in the first. Banco, intervening through the cupom cambial, essentially cheapened the “dollar” spread so that Brazilian banks that were already in a synthetic “dollar short” would add to it; which they clearly did outside of all these TIC numbers with the possible exception of the bank liability figure. In other words, in a purely financial maneuver, the Brazilian central bank made it exponentially worse so that when the next wave hit it would be both utterly devastating and almost completely obscured. Thus, an “unexpected” collapse.
In that respect, the TIC report provides an invaluable assistance into an otherwise hidden world. As noted above, the usual caveats still apply as what is reported and assembled isn’t the full picture, but given how the world has turned out so far in 2015 we can be reasonably assured that what we do have has been sufficient. The update for July provides still more evidence that the global “dollar” problem is purely bank-driven and that all the rest is the world trying as best it can to alleviate somewhat until forced into shocking and desperate measures just maintain working order, a step away from full chaos.