Store going out of business
lillisphotography/Getty Images

If you’re trying to figure out how the U.S. economy is doing these days, you probably feel like you’re riding on a seesaw.

One minute, things look like they’re heading up. U.S. employers created new positions at a healthy pace in July, adding 164,000 workers while the unemployment rate held at a near half-century low. Seconds later, however, it looks like there could be trouble. Days after President Donald Trump announced plans to slap additional tariffs on imports from China by Sept. 1, many economists started warning about the rising risks of a recession.

Most experts say there’s no need to panic just yet, but there are a number of datasets you may want to watch to help signal future economic shocks. They’re generally referred to as “indicators,” and they’re what experts read when trying to spot-check the health of the national economy.

Every week, a variety of these statistics are released. Though there’s no best indicator to follow, some are more widely watched than others. Here’s a breakdown of the indicators that consumers should keep an eye on, according to experts.

1. Yield curve

One of the most closely watched indicators of an impending recession is the “yield curve.”

A yield is simply the interest rate on a bond, or Treasury. These Treasuries have differing lengths of duration, known as their maturity. Some bonds last one month; some last 30 years. The curve, therefore, compares how those interest rates change over time.

Typically, the interest rate – how the U.S. government issuing these securities compensates investors for risks – is higher on a bond with a longer maturity.

[COMPARE: Best online savings accounts and rates]

“If you’re an investor, and you’re given the choice for investing for a month or investing for 10 years, you would say, ‘Listen, a lot more can go wrong in 10 years than it can in a month. I’m going to demand a higher interest rate, a higher yield,’” says Dan North, chief economist at Euler Hermes North America. “The yield curve is positive sloping – most of the time.”

When the yield curve inverts, however, the curve becomes downward sloping. This inversion means investors are demanding a higher yield for Treasuries of shorter durations. In this case, they think it’s riskier to hold their bond over the shorter-term.

Traders, economists and strategists alike watch this curve because it has a particular track record for preceding downturns. In a world where economic indicators come with a lag, most data is naturally backward looking. Inversions of the yield curve, however, have preceded recessions for the past 50 years, North says.

There are two main yield curves that investors follow for their predictive power: the 10-year and three-month Treasury rate, and the two-year and 10-year Treasury yield.

The 10-year, three-month spread inverted on March 22 for the first time since 2007, but it soon recovered. On May 23, however, the spread inverted again, and it’s stayed that way since (though briefly reverting on July 23). The two-year, 10-year curve, a spread that financial markets take even more seriously, inverted on Aug. 13 for the first time since 2007, with the two-year yield surpassing the 10-year yield by one basis point.

Meanwhile, the 30-year Treasury on Aug. 14 reached an all-time low of 2.053 percent, on fears about a sustained economic slowdown.

2. Confidence indexes

Even in economics, it matters how people feel. Case in point: The ongoing U.S.-China trade war. The conflict officially started in July 2018, when 25 percent tariffs took effect on $50 billion worth of goods. But since then, it’s escalated, with 25 percent tariffs now in place on an additional $200 billion worth of products.

The tit-for-tat nature of the trade war has firms feeling uncertain about the future. They’ve reportedly delayed investments and – in some cases – hiring, Federal Reserve Chairman Jerome Powell pointed out during the Fed’s post-meeting news conference on July 31.

“Trade tensions are having an effect on the U.S. economy. You see it now in the second quarter. You see weak investment, you see weak manufacturing,” Powell said. “With trade tensions — which do seem to be having a significant effect on financial market conditions and on the economy — they evolve in a different way, and we have to follow them.”

It’s worth following sentiment if you’re trying to predict what the economy will look like in the future, says Peter Donisanu, investment strategy analyst at the Wells Fargo Investment Institute.

“Sometimes a recession could be self-fulfilling. You build up so much pessimism about the economy that activity stops,” he says. “When you have businesses and consumers who feel less confident about the future, they have a lower propensity to spend. That’s when we start to see a contraction in growth.”

There’s no shortage of studies out there. For consumer confidence, the University of Michigan publishes a monthly consumer sentiment index with two updates: first a preliminary and then a final reading. The Conference Board, a research and business membership association, publishes its own gauge. Bloomberg also has its own consumer comfort index published weekly.

Meanwhile, the National Federation of Independent Businesses publishes a monthly survey showing sentiment among small firms. The Organization for Economic Cooperation (OECD) also publishes its own survey for businesses.

Consumer confidence remains historically elevated, though it tumbled in early 2019 as a result of the lengthy government shutdown. It’s since recovered and held stable. Meanwhile, the OECD’s business confidence index dipped below 100 for the first time since 2016, which indicates that businesses are starting to feel pessimistic about future performance.

But don’t fret about month-to-month fluctuations. It can be important if confidence slides, but mostly for the longer term, says Ryan Sweet, director of real-time economics at Moody’s Analytics.

“The relationship between confidence and spending is pretty loose in the short run,” Sweet says.

3. Employment Data

The day the monthly jobs report is released is often economists’ favorite day of the month.

During (what’s generally) the first Friday of each month, the Department of Labor publishes the broadest measure of the job market. The report contains a number of data points, including the percentage of the workforce that’s unemployed and the number of jobs each sector created.

It’s important to pay attention to these employment measures.

But there are more than just headline numbers worth following. Deeper within the report is what some economists say is a leading indicator: the number of hours worked.

“It’s a decent leading indicator. When the economy slows, businesses are getting worried about future sales,” Sweet says. “The first thing they cut are hours.”

Temporary help is another good measure, Sweet says. When businesses hire temporary workers, it could indicate that they’re not as confident about the future of the economy.

Both of those measures have remained fairly consistent, though it edged down to the lowest level since Sept. 2017 in the Labor Department’s most recent jobs report. Average weekly hours worked edged down in July to 34.3 from 34.4, where it had been since April.

The number of individuals who completed temporary jobs also edged up in July, at about 602,000 individuals, compared with 577,000 in June, the latest employment report showed. Over the past five months, that total has averaged out to about 579,000 individuals.

Sweet’s favorite, however, is jobless claims. This indicator provides a more down-to-the-minute view of the labor market, he says. It’s published every Thursday, compared with monthly for the Labor Department’s report.

“They’re fairly easy to count. It’s just the number of people applying for unemployment benefits,” Sweet says. “If something was wrong with the labor market, it would come up in weekly jobless claims.”

Initial jobless claims for the week that ended on Aug. 3 totaled 209,000, according to the Department of Labor. During the week that ended on April 13, however, claims fell to a near 50-year low.

[READ: 5 ways to help make your career recession-proof]

4. The Federal Reserve Bank of New York’s recession probability model

Based on the spread between the 10-year and three-month Treasury yields, the regional New York Fed compiles data about the likelihood of a recession in the year ahead. It’s updated typically at the beginning of each month, with the last release occurring on Aug. 2.

Currently, the index says there’s a 31.5 percent chance of a recession within the next 12 months. That’s the highest since before the start of the Great Recession and a warning sign for the U.S. economy, Donisanu says. That’s because the index doesn’t typically flash a yellow light until it reaches 30 percent.

But even though it has reached that threshold, it still wouldn’t be completely out of line with other economists’ estimates, who are starting to forecast that a recession could happen in 2020.

“Once we hit that 30 percent [level], it’s suggesting that a recession is still 12 months from now,” Donisanu says. “That would still be putting us through the middle of 2020.”

5. Leading Economic Index (LEI)

Most economic indicators are published with a lag, so it’s hard to find a forward-looking signal. In addition to consumer sentiment, the Conference Board also publishes an index composed of 10 datasets that predicts the direction of the global economy. Together, they make up the leading economic indicators index.

Some of the data included in this indicator are: Average weekly initial claims for unemployment insurance as well as the prices of 500 common stocks, according to the Conference Board.

Some economists track whether there have been three consecutive drops in the index, but it’s worth looking at the year-over-year change, Donisanu says.

“If we look at it in terms of when recessions happen, you don’t have a positive year-over-year change in the index,” he says. “That’s consistent going back through history, back to 1967.”

LEI decreased in June for the first time in 2019, but it’s not yet time to panic. Year-over-year growth hasn’t been negative, though it does appear to have leveled off. The index posted no change from April to May, and between March and April, the index only grew by 0.1 percent to 111.8.

6. Gross domestic product

A recession basically means that the economy isn’t growing — and the barometer of economic growth is a measure called gross domestic product (GDP). This means it’s worth monitoring this quarterly data because any signs of a faltering economy are bound to show up here.

But it’s also worth comparing gross domestic product with economists’ longer-run expectations, Donisanu says. This can help you know whether general fluctuations in GDP are any cause for concern, and it can also help you determine what the “output gap” is.

An output gap shows the difference between the actual output for the U.S. economy and what economists view as its maximum potential, and it can be either positive or negative. Real GDP, which shows growth adjusted for inflation, exceeded its potential in 2018 for the first time since 2007, according to a report from the Congressional Budget Office.

“When actual GDP is rising above its potential, we’ve historically seen these trends where the actual GDP rises above potential and it hits sort of a peak, rolls back over and falls below potential,” Donisanu says. “These are the cycles that ebb and flow over time.”

The U.S. economy grew by 3.1 percent in the first quarter of 2019 and expanded by 2.1 percent in the second quarter of 2019, according to the Department of Commerce. The Federal Reserve’s long-run growth potential for the U.S. economy is at 1.9 percent, according to its Summary of Economic Projections updated in June.

Bottom line: Don’t panic

Predicting a recession is “a fool’s game,” Donisanu says. Paying attention to these indicators helps, but there’s never a clear-cut signal.

“No one can ever say with such precision, ‘I know on the third quarter of 2020 a recession is going to start,’” Donisanu says. “We look at indicators, but even the best economists who can predict a recession with a high level of accuracy is low.”

The National Bureau for Economic Research’s Business Cycle Dating Committee is responsible for determining when a recession officially starts and ends. The last recession started in December 2007, the committee determined, but they didn’t officially declare this start date until a year later.

“The goal is not to be quick,” says Jeffrey Frankel, professor of capital formation and growth at Harvard University who’s on that committee. “Our assignment is to be as authoritative as we can. We’ve never had to revise a date, and it would create some problems if we did.”

Economists may even be scarred by the last downturn, Sweet says. They could potentially declare a recession prematurely.

“They may have a case of the yips,” he says. “They don’t want to miss two in a row.”

Still, when the inevitable downturn comes, it’s important to remember that not all recessions are as bad as the last one, he adds.

“That wasn’t a normal recession,” Sweet says. “The next one is going to be more of a garden variety.”

Learn more: