Petrodollar Puzzle: What Do Low Oil Prices Mean For U.S. Debt?

| About: iShares 20+ (TLT)

Summary

The dramatic plunge in crude prices catalyzed the demise of the petrodollar, the system that for years saw exporters recycle their oil proceeds into USD assets.

All else equal, this should drive Treasury yields higher - but all else is almost never equal.

Does the combination of a (relatively) hawkish Fed, Chinese reserve liquidation, and the continued dismantling of the petrodollar regime presage a Treasury sell-off?

Late last August, as China was drunkenly burning through its FX reserves following a yuan devaluation gone horribly awry, Deutsche Bank added a new term to the post-crisis lexicon: "quantitative tightening", or "QT."

The idea is simple: when central banks draw down their rainy day funds composed of USTs (NYSEARCA:TLT) and other core paper, they are effectively doing the opposite of quantitative easing. That is, they're selling DM government bonds. Proud of themselves, Deutsche Bank's analysts called their analysis "profound." Here are some excerpts from the profundity:

Declining FX reserves should place upward pressure on developed market yields given that the bulk of reserves are allocated to fixed income. A recent working paper by ECB staff shows that the increase in foreign holdings of euro area bonds from 2000 to mid-2006... is associated with a reduction of euro area long-term interest rates by about 1.55 percentage points, in line with the estimated impact on US Treasury yields by other studies.

On the short-term impact, one recent paper estimates that 'if foreign official inflows into U.S. Treasuries were to decrease in a given month by $100 billion, 5- year Treasury rates would rise by about 40-60 basis points in the short run', consistent with our estimates above. China and oil exporting countries played an important role in these flows.'

In other words, all else equal, China's reserve liquidation along with the reversal of capital flows in oil exporting countries should have put quite a bit of upward pressure on UST yields considering the concentration of reserve managers on USD assets.

Of course all else is never equal and between Janet Yellen's dovish relent in September and episodic safe haven flows, the market never really saw the kind of spike in yields that China's massive reserve liquidation seemed to presage. In fact, just before equities bounced off the lows on February 11, the yield on the 10-year was all the way down to 165. Kuroda and Draghi have helped - with so much European and Japanese paper trading negative, investors are happy to buy USTs for the comparatively juicy yield.

I bring all of this up because on Monday, the Treasury Department revealed the size of Saudi Arabia's UST holdings for the first time after someone over at Bloomberg decided to submit a Freedom-of-Information Act request.

Back in January, Bloomberg ran a piece that carried the headline "Saudi Arabia's Secret Holdings Of US Debt Are Suddenly A Big Deal." It was a good read and I'm reasonably sure Treasury knew an FOIA request was forthcoming once the article ran. "It's a secret of the vast U.S. Treasury market, a holdover from an age of oil shortages and mighty petrodollars: Just how much of America's debt does Saudi Arabia own?," Bloomberg asked. "That question -- unanswered since the 1970s, under an unusual blackout by the U.S. Treasury Department -- has come to the fore as Saudi Arabia is pressured by plunging oil prices and costly wars in the Middle East," the piece continued.

Unsurprisingly, Bloomberg cited Deutsche Bank. "Saudi Arabia's moves have drawn scrutiny, particularly as other central banks in emerging markets sell Treasuries to raise cash in defense of their currencies," they wrote, adding that "according to Deutsche Bank, selling by foreign central banks since March has added 0.3 percentage point to yields on 10-year Treasuries."

Well, as it turns out, the "official" data suggests the Saudis own a comparatively small amount of US debt. Riyadh is sitting on just $116.8 billion in USTs according to the data released this week, far below China's $1.3 trillion stash, and Japan's $1.1 trillion in holdings. Here's the breakdown of producers' UST holdings:

click to enlarge) Click to enlarge

(Table: Bloomberg)

As Bloomberg notes, the figures raise more questions than they answer. Here are some key excerpts from their piece:

In a sign that Treasury's figures on the kingdom's ownership fall short of the full tally, the New York Times reported last month that Saudi officials threatened to sell $750 billion of Treasuries and other assets in the U.S. if Congress enacts a bill allowing the monarchy to be held responsible in American courts for any role in the Sept. 11, 2001, terror attacks.

The U.S. started releasing data on foreign ownership of Treasuries in 1974. Since then, the Treasury's policy had been to not disclose Saudi holdings.

The special arrangement was a product of the 1973 oil shock following the Arab embargo. It's among concessions that U.S. administrations made over the years to maintain America's strategic relationship with the Saudi royal family and access to the kingdom's oil reserves.

In the past year, Saudi Arabia burned through 16 percent of its foreign-exchange reserves to plug its biggest budget shortfall in a quarter-century, according to data from the kingdom's central bank. The signs of strain are prompting concern over Saudi Arabia's potential influence on the world's largest and most important bond market.

In short, there's no telling what the "real" figure is. As Bloomberg goes on to point out, some nations (ahem.. China) hold USD assets in foreign custodial accounts and frankly, the headline number for Saudi UST holdings looks too low to be true.

But even if reserve liquidation across EM and oil producing economies hasn't yet had a pronounced effect on DM bond yields, and even if the Saudis don't own as much US paper as foreign minister Adel al-Jubeir suggested they hold last month when speaking to President Obama about the 9/11 issue and/or as much as many on the Street suspected, the reversal of petrodollar flows has real long-term implications for fixed income and especially for USTs.

As I discussed last week, the Gulf producers control an enormous amount of foreign assets between them. Qatar, Bahrain, Saudi Arabia, Kuwait, Oman, and the UAE hold something like $2.5 trillion in combined reserves.

(Chart: Deutsche Bank)

According to Deutsche Bank's estimates, only Kuwait can break even (fiscally speaking) at current prices for crude.

That means everyone else will need to fund their deficits by either tapping debt markets or liquidating reserves.

Throw in the enormous pressure exerted by the region's various USD pegs on top of the cost associated with funding wars in Yemen and Syria, and you can expect the petrodollar reversal to persist for the foreseeable future.

And while everyone knows that the death of the petrodollar is important, the trick is to quantify the damage in terms of rising yields. If Wednesday's Fed Minutes are any indication, the FOMC does intend to move sooner rather than later. And eventually, the ECB and the BoJ will have to attempt to normalize policy as well. If "lower for longer" is indeed the new normal for crude, the question becomes this: what happens when the reversal of petrodollar flows meets a continual drawdown in Chinese reserves and a Fed tightening cycle?

Goldman - who earlier this week moved forward their forecasts for higher oil prices - is out with a fresh take on the effect low crude is having on capital flows and ultimately, on yields.

Capital exports from oil producing countries (ex-advanced economies like the US and Norway) averaged $300 billion per year from 2006 to 2014, the bank says, before noting that last year, the dynamic reversed as crude plunged. Here's a look at the relationship between crude prices and petrodollar flows:

(Chart: Goldman)

And here's a look at 12-month SAR forwards (in the simplest possible terms, this is a measure of how much pressure there is on the Saudi riyal's dollar peg):

click to enlarge) Click to enlarge

(Chart: Bloomberg)

Moving to specifics, Goldman goes on to quantify the level at which oil producing economies as a group swing from CA deficits to surpluses; in other words, the price level at which petrodollars are available for investment in things like USTs.

The bank also explains what I've outlined at length recently: Saudi Arabia's position is weakened by the fact that the kingdom spends heavily on social welfare. The bank also touches on how expensive it is to maintain the USD pegs:

For the major EM oil producers, our capital flow breakeven estimates range from a low of around $15-20pb for Russia and Kuwait to a high of around $45-$50pb in the case of Saudi Arabia and Nigeria.

It is worth emphasizing that the breakeven level of oil prices for petrodollar capital flows is a very different concept from the breakeven price of oil production costs in that economy. For instance, most estimates suggest that the breakeven cost of oil production in Saudi Arabia is relatively low. However, the level of oil prices at which Saudi Arabia has historically started to run a current account surplus and capital outflows is relatively high, because the level of public sector service provision in that economy has been predicated on the assumption that high oil prices would be sustained. This underlines the need for public sector reform in that economy and has been a driving force behind the political changes that are underway. It has also led to speculation that Saudi Arabia and other GCC countries may decide to devalue their exchange rate pegs.

(Chart: Goldman)

Next, Goldman explains the critical link between petrodollar flows, global savings, and rates. Essentially, the idea is that when oil prices fall, wealth shifts from economies that are inclined to save (i.e. the producers) to economies that are inclined to spend (think about the effect lower gas prices purportedly has on Americans' level of disposable income).

When the aggregate level of global savings falls, bond yields should of course be expected to rise. Here's Goldman again:

In Exhibit 5 we display the correlation between oil prices and global savings.

One would expect a reduction in the supply of savings - represented by a northwesterly shift of the savings schedule in Exhibit 6 - to result in an increase in global real bond yields, all else equal.

(Charts: Goldman)

Right. But I'll say it again: all else is never equal.

Goldman goes on to list a number of factors that have suppressed yields even in the face of China's epic reserve drawdown and the reversal of the petrodollar dynamic. Although they mention central bank purchases (by the BoJ and ECB) as a contributing factor, they nevertheless contend that there are other, more salient explanations.

I would tend to disagree. It seems entirely likely that this is a tug of war between the DM central banks that are still easing (which is almost all of them), the unwind of the petrodollar system, and EM reserve liquidation. The latter two factors are exerting upward pressure on yields while the former is catalyzing spread compression.

Draghi is buying €80 billion in bonds a month and Kuroda targets JPY80 trillion in JGB purchases per year. Not only does that exert downward pressure on EGB and JGB yields, it also drives investors into USTs, pushing US yields lower in the process.

So will central banks succeed in offsetting the effect of the petrodollar unwind and China's massive UST liquidation effort? Or will the Fed's next hike be just enough to tip the scales, causing long-end rates to rise and dealing a devastating blow to those who have piled into fixed income at historically low yields?

Time will tell.

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.