Why Higher Oil Results In Higher Yields: A Surprise Explanation From JPMorgan

When one considers the relationship between crude oil prices and 10Y yields, conventional wisdom traditionally looks at the correlation between 10Y Breakevens – i.e., the inflation expectation component of interest rates while when combined with real rates make up nominal 10Y yields – and the price of crude, which as shown in the chart below, have almost overlapping correlation.

Yet according to a report by JPMorgan’s “flows and liquidity” wizard, Nick Panagirtzoglou, this is one case where correlation most certainly does not imply causation.

Writing over the weekend, Panagirtzoglou observes that, as shown in the chart below, the sharp increase in oil prices over the past week has been accompanied by a rise in the 10y UST yield to a new high, with the latter tracking the former closely.

However, in turning the tables on what is one of most conventional explanations in the market, the JPM analyst writes that “while the temptation is to associate this relationship with inflation expectations”, i.e., Breakevens, he thinks this is “not credible explanation” because he believes that most of the recent rise in bond yields has been due to real yields rising rather than inflation breakevens (although looking at the chart up top, we don’t agree with this statement). Instead, JPMorgan believes that a “more credible” explanation is that “bond markets are actually reacting to an anticipated flow shift caused by the rise in oil prices.”

So what, according to the JPM analyst is the flow shift resulting from higher oil that bond markets are currently anticipating? And why is his explanation divergent from the other variant perception, namely that the plunge in oil prices in the 2014-2017 period prompted a liquidation of “safe assets” by oil exporters and EMs?

Read on for his full explanation, which while provocative may need a little more work to be convincing.

We argued before that the shifts in flows and incomes resulting from higher oil prices are creating a bearish flow for bonds, effectively reversing the previous bullish flow shifts seen between 2014 and 2016. Oil consumers had spent $3.4tr on crude and related products in 2014 with an average oil price of $100 per barrel, and in 2016 they had spent less than half of that, i.e. $1.6tr with an average oil price of $45 per barrel. In other words there was a massive $1.8tr or 2.2% of global GDP income transfer between oil consumers and oil producers between 2014 and 2016. These income transfers have been reversing over the past two years as the average oil price rose from $45 during 2016 to $55 during 2017 and to $80 currently in 2018. If we assume an average oil price of $80 for the remainder of 2018, close to the current level, it would mean that almost two thirds of the previous 2014/2016 income transfer would have been reversed over the past two years i.e. between 2016 and 2018, with most of this reversal taking place between 2017 and 2018. How are these oil income shifts creating a bearish flow for bonds?

The answer is via oil consumers. The massive boost to income for oil consumers between 2014 and 2016 had created a supportive savings flow into fixed income. In particular we believe that a decent part of the previous $1.8tr income windfall seen between 2014 and 2016, equally split between the residential sector, the industrial sector and the transportation sector, was likely saved. These savings most likely took the form of bank deposits which eventually supported bond markets via the banking system deploying these excess deposits into government bond markets. As a result, the oil income windfall to households and oil consuming industries had likely created a bullish flow into fixed income between 2014 and 2016, bigger in size than the bearish fixed income flows resulting from the decumulation of SWF/FX reserves of oil exporting countries during these two years.

Two things to note in the paragraph above: while the JPM explanation that oil consumers benefited from low oil prices and saved money that otherwise would have been spent on gas and other forms of energy, and invested it in bonds, what we find curious is JPM’s explanation of how increased deposits were converted by banks into bonds, which suggests once again that QE is not a loan-creating mechanism at all, but one which intermediates in the markets, and prompts banks to directly boost asset prices.

The other point is JPM’s assertion that savings by oil-importing consumers more than offset the “bearish fixed income flows resulting from the decumulation of SWF/FX reserves of oil exporting countries during these two years.” Here we disagree for one main reason: as we showed before, much of the “savings” from lower oil prices were never actually saved – especially with the US savings rate recently dropping to all time lows, but rather spent on deleveraging, and various other discretionary purchases. And yet JPM is right that at the deflationary peak in 2015 and early 2016, yields were likewise at the lows, suggesting that indeed the “bearish” flow from oil exporters was being offset somewhere. If JPM is right, this would be a major milestone in the explanation of who holds the trump cards during the next oil-price driven recessionary crash.

So going back to the JPM narrative, Panagirtzoglou then looks at the events since oil prices bottomed in 2016, and notes that “the opposite has been happening since 2016.”

These positive bond flow dynamics emanating from oil consumers started reversing post 2016 as oil prices started rising and as the previous income windfall started eroding. Again, factoring in a rise in oil price from $45 in 2016 to $55 in 2017 and $80 in 2018 implies an income squeeze of $1.1tr for oil consumers, a decent portion of which likely taking the form of reduced flow of savings. In turn, this exerts downward pressure on overall bond demand, and upward pressure on bond yields.


While this is an interesting hypothesis, its biggest implications, as pointed out earlier, are for the next oil-price swoon, because if JPM is right it would mean that for all the fears that oil exporters and EMs could launch a bond market crisis by being force to liquidate their holdings, it is the “savers” on the other side, the consumers in oil-importing nations, that would more than offset the damage from this forced liquidation, resulting in a deflationary slump which parallels what is observed in the commodity space. And since algos will again confuse tumbling yields for a disinflationary signal, it also means that the next oil plunge will have a dire effect on stocks, especially if cross-stock and cross-asset correlations remains as elevated as they have been in recent months.


Why Higher Oil Results In Higher Yields: A Surprise Explanation From JPMorgan