Last week, I asked a question that's been debated ad nauseum over the past couple of years: "does anyone want America's debt?"
Obviously, that's a rhetorical question. The answer, of course, is "yes." Demand, I noted, probably shouldn't be at the top of your worries list when it comes to a long US Treasury (NYSEARCA:TLT) position.
But as usual, it's the nuance that matters.
The backdrop against which the debate is set revolves around the extent to which China and Saudi Arabia have trimmed their FX reserves in the wake of the RMB devaluation and lower crude prices, respectively.
In China, Beijing is attempting to manage the RMB's downward trajectory and that's an expensive effort. The trick is to guide the currency lower while intervening in the spot market to ensure it doesn't fall too far, too fast. Intervention means liquidating core paper. Since the initial ("one-off") August 11, 2015 devaluation, the PBoC has seen its FX reserves fall from a peak of around $4 trillion to just under $3 trillion with the latter being seen by some observers as a psychologically important level.
As for Saudi Arabia, plunging crude prices (a largely self-inflicted wound) blew a hole in the budget and before you knew it, the kingdom was running a fiscal deficit that at one point rose to some 16% of GDP. As it turns out, it's expensive to subsidize the lives of everyday Saudis and it's also expensive to maintain the riyal's dollar peg. In order to stanch the SAMA bleed, the Saudis cut subsidies and tapped debt markets for $17.5 billion last October.
But it's not just EM and GCC countries that are selling US debt. As I documented in the post linked above, Japan has been selling en masse of late. In fact, Japan sold the most US paper in four years during December.
Specifically, Japanese private investors dumped some $20 billion in December, the most since May of 2013. Here's a fun visual with giant red arrows to help you along:
As I noted last week, Japanese banks have been big sellers of US debt since the election. Here's the breakdown (note the purchase cost in the right pane):
(Charts: Morgan Stanley)
So clearly, the motivation here is to cut losses as rates rise. "Banks have a known tendency to either take profits or cut losses in the event of significant market movements," Morgan Stanley wrote last week.
Bloomberg tells a similar story in a piece out Monday. To wit:
Last quarter, Japanese investors who hedged all their dollar exposure in Treasuries suffered a 4.7 percent loss -- the biggest in at least three decades, data from Bank of America showed. The same thing happened in Europe, where record currency-hedged losses also stung euro-based buyers.
"It was a deer in the headlights moment," said Zoltan Pozsar, a research analyst at Credit Suisse.
Combined with the unpredictability of Trump's tweet storms, interest-rate increases in the U.S. could further sap overseas demand. Mark Dowding, who helps oversees about $50 billion as co-head of investment-grade debt at BlueBay Asset Management in London, says the firm has already moved to insulate itself from further losses due to higher rates.
As Bloomberg goes on to note, even when hedging costs are taken into account, the yield pick up in Treasurys over JGBs is notable:
The fact that Japanese investors are selling despite this pickup speaks volumes about their outlook for US rates.
In any case, you'll note that all of this has implications for the dollar (NYSEARCA:UUP). Needless to say, the higher the hedging cost, the more likely foreign investors will be to run unhedged UST positions and the more unhedged positions there are, the more upward pressure on the greenback.
Now, remember last year when I talked incessantly (one of the very early posts is here) about the seemingly esoteric cross-currency basis market and how important it was if you want to get a read on the global dollar funding crunch? Well, when I told you that was important, I wasn't making it up.
As the demand for USD assets (i.e. Treasurys) rises, the global dollar funding crunch accelerates. One way to gauge this is by observing the extent to which the cross-currency basis is negative. Prior to the crisis, basis levels hovered around zero. Post-crisis, it's a different story.
Basically we're just talking about collateralized borrowing here. I want dollars, you've got dollars. I'm the borrower, you're the lender. I pay you LIBOR and you pay me EURIBOR (or Yen LIBOR, or whatever), plus some points on the currency collateral. When those points are negative, it's indicative of increased dollar demand.
Ok. So what do you imagine has happened to cross-currency bases in an environment where the demand for US Treasurys has dissipated? That's right, they've turned less negative. In fact, they've tightened close to their two-year averages:
Here's BofAML with some color (my highlights):
One of the primary drivers for the recent basis narrowing has been a significant decrease in overseas demand for USD assets. This decline is evidenced by reduced Japanese and European overseas investment demand, likely related to (1) deleveraging ahead of Japanese fiscal year end (2) increased uncertainty regarding US fiscal policy and the expectation for higher US interest rates.
Note that everything said above about widening cross-currency basis levels also applies in reverse. That is, as demand for JGBs picks up, the USDJPY cross-currency basis will narrow. Here's BofAML again (my highlights):
Another contributor to the tightening of JPY bases has been foreign purchases of JGBs. Foreigners continue to purchase JGBs at an aggressive pace (Chart 11), with the lowervol and higher yield pick-up on a fully hedged basis proving attractive to overseas investors (Chart 12). Recall, when foreign investors purchase Japanese assets they typically fully currency hedge their exposure which provides USD to funding markets and pressures the JPY cross currency basis tighter. JSDA data shows that net purchases by foreign investors have grown tremendously over recent quarters, while increasing JGB purchases by European investors can also been seen in Chart 6 above (yellow bars).
But here's the thing: the basis narrowing has probably overshot. Or at least that's BofAML's contention.
Part of the rationale for that contention comes back to money market reform. Foreign banks can't raise USD funding as quickly as they once could given the outflow from prime money market funds (for more, please see here) and commensurate decline in demand for commercial paper and time deposits. So ultimately, foreign investors like Japanese banks will have to come back to the cross-currency basis market eventually.
As for EURUSD basis, well, the reasons why it should blow back out should be readily apparent, but in case they aren't, here's BofAML one last time:
We have more conviction in our expectation that the short-dated EURUSD basis should widen due to (1) emerging political risks (2) upcoming ECB liquidity supplying operations (3) potential repatriation flows (4) expectations for EUR repo market squeezes.
So in the end, what you should take away from this is that the global dollar funding crunch has been alleviated somewhat of late thanks to flagging demand for US paper (i.e. Treasurys).
That said, recall what I said above about pre-crisis versus post-crisis levels in the cross-currency basis market. The basis should be zero (just like in theory, the cash-CDS basis in credit markets should be zero). But you can bet it won't be and if you believe the world is going to continue to demand dollars, you can bet on these markets widening back out materially going forward.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.