Fickle Markets Flash Recession Signs as Economists Say Keep Calm
(Bloomberg) -- Market-based recession indicators are flashing red. To some economists, they’re better left ignored.
At the moment that’s easier said than done. Among a slew of gauges maintained by JPMorgan Chase & Co., one based on stocks and credit spreads puts the probability of a recession in the next 12 months at 50 percent, though it was at 70 percent two weeks ago. Another, based on the gap between long-term and short-term Treasury yields -- the so-called yield curve -- puts the figure at 46 percent, according to economist Jesse Edgerton.
When it comes to predicting recessions, market-based models carry one clear advantage: Compared with hard economic data, they’re more forward-looking. But not only can markets get it wrong, they’re also prone to rapid reversal, undermining their reliability as the harbingers of doom.
“Financial markets are efficient, and they can be powerful predictors of future economic conditions, including recessions,” said Joseph Davis, chief economist at Vanguard Group, the mutual-fund giant. “That doesn’t mean they’re always accurate.”
Read More: Slowdown or Meltdown? Reading the Tea Leaves on Global Growth
Such caution may be warranted as the U.S. trade dispute with China and the partial U.S. government shutdown drag on. While each looms as a threat to economic activity that has been priced in to financial markets and affected measures of business confidence, neither has seriously damaged U.S. output, and could disappear before that happens.
“If we can get a resolution on just one of these fronts, some of the recession risks should subside,” Davis said.
Equity markets, in particular, are notoriously alarmist. As the late economist Paul Samuelson famously wrote in 1966, stocks had “predicted nine out of the last five recessions.” Its mistakes, he added, “were beauties.”
To its credit, the yield curve has been more reliable. The spread between three-month bills and 10-year Treasuries has inverted before each of the past seven recessions. But the usefulness of that signal is muted because the timing of subsequent recessions is so varied and unpredictable.
What Our Economists SayEconomic data is backward looking, market data is forward looking. For that reason alone, we should pay close attention to what markets are saying about recession risks. At the same time, volatility and factors like the Fed balance-sheet unwind can make market signals difficult to read. At the present moment, we think the message from the economic data -- slowdown not meltdown -- is the right one.
-- Tom Orlik, chief economist, Bloomberg Economics
According to Roberto Perli, an economist at Cornerstone Macro LLC in Washington, the time lag from inversion to recession has ranged from nine months for the 1981-82 downturn, to 34 months for the 2001 recession, with no discernible pattern.
The first lesson is to avoid relying on any single indicator. But even composite measures are, at the moment, attracting skepticism. Perli’s own model based on stocks, corporate-bond spreads and various portions of the Treasury yield curve warns of a 56 percent chance of recession within a year. And yet, he advised his clients in a note Tuesday that “these models are best ignored.”
That’s because Perli pins much of the recent market woes on angst over China, which he thinks will eventually ease as monetary policy in China adjusts and prevents a hard landing.
“The markets are expressing worries about the Chinese economy and its potential repercussions on the U.S. economy,” wrote Perli, a former Fed official. “There doesn’t seem to be much specific to the U.S. economy that seems likely to go seriously wrong over the next few quarters.”
While Perli’s optimism over China can be debated, it’s clearer that a potential recession in China is weighing on U.S. financial markets before it’s baked into the economic outlook. In other words, it’s a risk factor but not a hazard that has been struck.
As for the hard economic data, that may be slower than markets to reveal a deteriorating outlook, but less likely to give firms and families a head-fake on what’s coming.
“You cannot get the equivalent of a thousand-point drop in the real economy in 24 hours, but you can in the financial markets,” said Bill Dunkelberg, chief economist for the National Federation of Independent Business. “So, if the model you’re looking at is dominated by financial numbers, there’s a much higher chance you can get a major reversal.”
It’s somewhat comforting, then, that the hard data are signaling a slowdown in growth, but no recession. Edgerton said JPMorgan’s “preferred model,” based on economic data, puts the probability of a recession in the next year at 41 percent.
That compares with the 25 percent median chance of a slump seen by economists in a Bloomberg survey earlier this month.
Vanguard’s Davis has a model which blends economic data and market measures and has spit out a 35 percent probability. “We’ll still have growth scares,” he said. “But fate hasn’t been set for 2019.”
©2019 Bloomberg L.P.