It seems as though financial descent looms just around the corner.

U.S. stock and bond markets are both moving toward unpleasant milestones, foreign investors are eyeing U.S. government debt warily, and the unofficial moratorium on tossing the word “recession” into headlines has lifted.

The current recession focus hinges on a humble, abstract measure known as the yield curve. The wider world only notices it once or twice each decade, when folks discover anew that it predicts recessions more accurately than almost anything else.

When charted, the yield curve is a literal curve. It shows the return on U.S. government debt of different maturity lengths — everything from one-month Treasury bills to 10-year Treasury notes to 30-year Treasury bonds.

In a healthy yield curve, that long-term debt has higher returns than the short-term debt. That basically reflects that investors can be more certain about what prices and economic growth will look like in the short term (and the government’s ability to pay back debts) than in the long term. When investors get jumpy about the near term, however, that curve flips over like a capsized turtle. This is the dreaded “inverted yield curve.” Right now, the curve is flattening, but it hasn’t yet inverted.

You can see that below. The teal line is “flattening” below where the yield curve was a year ago. (In December 2006, a year before a historic recession, there were already signs of duress in the markets, and the yield curve was already quite inverted.)

An inverted yield curve has preceded every recession since 1960. In all that time, there’s been only one false alarm, in the mid-1960s.

Right now, the yield curve is in the teal zone below. The yield on the 10-year Treasury note is within less than a percentage point of the yield on the three-month bill, according to a popular measure from the Federal Reserve Bank of New York. When it hits the brown zone — full inversion — there will be cause for concern.

For now, there's wiggle room. The yield curve doesn’t always invert when it gets this flat. In the 1990s, during the longest economic expansion on record, the yield curve dropped to its current level or below without inverting on several occasions. When it did invert in July 2000, recession was less than a year away.

Which reminds us: Even when it turns on us, the yield curve doesn’t signal immediate doom. In the past 60 years, inversions have happened anywhere from five (1959) to 17 (2006) months before a downturn. And the false-alarm inversion in 1966 happened a full four years before a true recession arrived.

Even if the yield curve inverts and the economy deteriorates as rapidly as it did in 1959, recession is eight months away — and that’s an outlier. On average, it’s taken 16 months to go from the current level to inversion, and about 11 months to go from inversion to recession. Thus, it seems safer to assume the economy will keep growing for at least another year, at which point this expansion will catch the 1990s as the longest run on record.

But what about the stock market?

The Standard & Poor's 500-stock index is barely positive on the year, and the Dow Jones industrial average has lost ground since the end of 2017. The markets are in the doldrums and, once again, flirting with correction territory.

Correction is, of course, markets jargon for “down 10 percent.” But unlike an inverted yield curve, corrections aren’t reason to panic. They happen too frequently to accurately predict recessions.

Since 1960, corrections in the S&P 500, as defined by Yardeni Research, have occurred an average of 3.25 years before a recession. But that average is almost meaningless, given the irregularity and frequency of the events involved. The market has been known to venture into correction territory four or five times during the course of an expansion.

Here’s the chart from above again. This time, focus on the red blips at the bottom. The only lesson? The stock market isn’t great during recessions. But sometimes it’s also not great outside of recessions.

Are foreign countries worried about U.S. debt?

The share of U.S. debt owned by foreign banks, governments and other investors has plunged almost three percentage points between October and April, the most recent month for which data is available.

The drop has fueled concerns of a “Trump dump,” in which foreign investors have been turned off by the president’s bombastic approach to politics and economics. But numbers don’t back them up.

“There’s been a lot of media commentary suggesting that foreigners are stepping back from U.S. Treasurys, and in some cases — particularly on Bloomberg — it’s been laid at Trump’s feet,” said Benn Steil, director of international economics at the Council on Foreign Relations. “But … it’s just not borne out in the data.”

As recently as 2015, foreigners owned about half of all U.S. Treasury debt. Their share has withered to just 41.6 percent, but it was falling at the current rate during 2015 and 2016 as well — long before Trump took power. In fact, foreign holdings actually climbed for the first half of his time in office so far. We’ll zoom out soon, but here’s the tight view:

According to an analysis by Steil and his colleague Benjamin Della Rocca, the vast majority of that drop can be attributed to just two countries, China and Japan. The East Asian nations have been shedding U.S. debt at a greater clip than the rest of foreign Treasury holders, but they’ve each got reasonable, Trump-agnostic reasons for doing so.

China’s Treasury holdings are falling not because of any sort of antipathy toward Trump but because the country has grown less aggressive about hoarding dollar debt to artificially devalue its currency.

Japan, for its part, has effectively been swapping its Treasurys for U.S.-backed mortgage debt — a move that doesn’t seem to demonstrate a lack of faith in the U.S. government. Japanese buyers may also be deterred by the rising cost of hedging dollar-yen investments.

For now other investors, many of them based in the United States, have picked up the slack. China’s huge Treasury stockpile has not given it much leverage in recent trade negotiations for the same reason — if China sells its assets at a cheap price for reasons not related to economic fundamentals, then others will be more than willing to take advantage of the bargain.

But Americans shouldn’t be complacent just because foreigners aren’t dumping Treasurys quite yet. The United States will need to issue a huge amount of debt to cover both a substantial tax cut and large spending increases, which creates a combustible situation. It hasn’t been a problem yet thanks to low interest rates, Steil said, “but if that interest rate were to spike, we could have a very significant problem.”

“That could particularly be the case if there’s a significant economic slowdown,” Steil said. “And a trade war could no doubt produce that.”

Correction: A previous version of this story misspelled Benn Steil's last name in some references.