Can an Oil Price Shock Derail the US Economy?

April 15, 2026 – Welcome back to another post. This time, I want to do another economic update, which many readers seem to enjoy. Today’s topic is whether the recent oil price shock, driven by Middle East geopolitical uncertainty, poses a threat to the U.S. economy. That risk is certainly pushed by the news and also reposted by some folks in the personal finance sphere. For example, the recent post on The Motley Fool, claiming their Moody’s AI model (which has never been wrong for 80 years), almost certainly predicts a recession over the next 12 months. What to make of that? After all, the Fool/Moody’s model is AI, so it must be right, right? They wouldn’t publish any clickbait, right? Well, not so fast! There are many reasons to feel much more relaxed about the recent oil price spike. Let’s take a look…

Basic business cycle stats

Before we even get into the mechanics of oil price shocks, let’s look at some basic business cycle facts and stats (all referring to data 01/1926-03/2026):

• Fun fact: In the U.S., recessions vs. expansions are dated by the National Bureau of Economic Research, which considers a range of monthly indicators. The U.S. approach to business cycle dating differs from the naive approach of “two consecutive quarters of negative GDP growth.” Also, keep in mind that the turning points are often declared with long lags and the benefit of hindsight. For example, the NBER called the start of the March 2001 recession in November 2001. By that time, the recession had just ended. Also, the 11/2001 end of the recession wasn’t made official by the NBER until July 17, 2003. Thanks for telling us!
• Since 1926, the US economy has gone through 16 recessions. The most recent one was the pandemic recession in 2020.
• The average recession lasted 12.6 months. Recession lengths ranged from just 2 months in 2020 to 43 months between 1929 and 1933. Expansions lasted 62.5 months on average. The longest expansion on record was from June 2009 to February 2020, 128 months!
• The unconditional probability of a recession was one-sixth, i.e., 16.7%. In 83.3% of months, we were in an economic expansion.
• There is “stickiness” or “serial correlation” in business cycle states, i.e., if we were in an expansion in one month, it’s highly likely (98.4%) to continue. Thus, there is only a 1.6% chance of slipping into a recession. Recessions are also “sticky,” albeit slightly less than expansions, i.e., if you were in a recession in month X, you’re likely to stay in a recession (92% probability) and escape only 8% of the time.
• We can also consider longer horizons; for example, if we’re currently in an expansion, there is a roughly 19% probability that we’ll be in a recession at least once over the subsequent twelve months.

So, why is everyone so worried about the risk of an oil shock to the economy? That brings me to the next point…

Oil Price Shocks often precede recessions

The picture that has some economists nervous is the one below. I plot the real GDP time series in the top panel and the (nominal) WTI crude oil price in the bottom. Indeed, almost every U.S. recession since the early 1970s saw an oil price spike prior to the business cycle turning point. The prominent exception is the pandemic recession, which is understandably not the result of an oil price shock. Quite the opposite: the temporary shutdown substantially lowered the WTI price.

So, should we worry about the correlation between oil price shocks and the risk of a recession? Here are a few reasons to relax…

Why I believe the oil price shock recession risk is exaggerated

1: The Nominal Oil price is misleading

True, $100 per barrel of oil sounds scary. Almost $120 in early April sounds even scarier. But to compare oil prices historically, we need to – you guessed it – adjust historical oil prices by CPI inflation to get a true apples-to-apples comparison. Suddenly, the $100 per barrel looks a bit more benign. More like a “diet” oil price shock. The 1980 oil price shock sent prices North of $160, adjusted for inflation, and the 2008 oil price peak was over $210. Also note the historical oil price spikes in 2011 ($168) and 2022 ($130) that did not cause a recession.

2: We are now much less reliant on oil consumption

Adjusting the 1970s oil price for inflation is only half the story. In fact, it’s probably much less than half. That’s because the U.S. economy is now much less reliant on oil consumption. I can demonstrate this in at least two different ways. Please see the chart below. First, in the top chart, I plot real GDP per barrel of oil consumption. Compared with the 1970s, today we can produce more than three times as much real GDP per barrel of oil. So, U.S. GDP growth should now feel about 70% less impact than in the 1970s.

In the chart at the bottom, I calculate the dollar value of energy consumption per $100 of GDP generation. This gives us a better tool for comparing the magnitude of oil price shocks over time. For example, in the 1970s and 80s, the oil input in $100 worth of GDP grew from under $2 to almost $9. This $7 increase had a major impact on productivity and the U.S. economy. In comparison, the current episode has a roughly $1 increase in the energy consumption share. But even this calculation vastly exaggerates the effect of energy prices, because it only considers gross oil consumption. This brings me to the next point…

3: We rely less on oil imports

Finally, looking only at oil consumption is not the full picture either. True, every (net) consumer of oil will feel the pinch of higher oil prices and might have to cut back on other consumption. But what about the oil producers, especially the U.S. domestic oil producers? Over the last few years, we’ve seen a remarkable renaissance of U.S. domestic energy production. The U.S. Department of Energy has data on crude oil consumption and domestic production, and in CY 2024, we’ve reached essentially crude oil self-sufficiency, i.e., domestic production has caught up with consumption.

So, higher oil prices likely don’t pose an aggregate risk to the economy. For every net energy consumer who needs to tighten the belt due to higher prices, there is another person who now enjoys a windfall profit and can go on a consumption (or investment) spree. Higher oil prices will generate winners and losers and pose more of a distributional rather than aggregate issue. Of course, some economists will still worry. The distribution effect could still cause an aggregate adverse effect on GDP if you subscribe to the old-style Keynesian prescriptions of multiplier models and claim that the oil price shock losers (poor people) have a higher consumption propensity than the winners (rich oil barons). But that’s an outdated way of thinking; last time I checked, GDP counts not just consumption but also investments. Specifically, if folks pocket some extra money from energy investments, they might reinvest that cash in more energy projects, which will certainly be counted in our GDP numbers.

Oil Price Shocks and the Macroeconomy: Academic Research

Without getting too geeky, I want to point the interested readers to some of my academic research from way back when I actually contributed something to the economics profession. I published three peer-reviewed academic journal articles on the topic of oil/energy price shocks and their effects on the macroeconomy:

• DHAWAN, R. and JESKE, K. (2008), Energy Price Shocks and the Macroeconomy: The Role of Consumer Durables. Journal of Money, Credit and Banking, 40: 1357-1377. https://doi.org/10.1111/j.1538-4616.2008.00163.x
• DHAWAN, R. and JESKE, K. (2008), What determines the output drop after an energy price increase: Household or firm energy share? Economics Letters, 101: 202-205. https://doi.org/10.1016/j.econlet.2008.08.011
• DHAWAN, R., JESKE, K. and SILOS, P. (2009), Productivity, energy prices and the great moderation: A new link. Review of Economic Dynamics, 33: 715-724. https://doi.org/10.1016/j.red.2009.07.001

The most relevant for today’s discussion is the third paper with my friends Rajeev and Pedro, in which we show that there was a structural break in the interaction and correlation between energy price shocks and productivity shocks. Prior to 1982, energy price and productivity shocks were highly correlated, but this correlation has since disappeared. We showed this with a pretty neat and innovative (at least back in 2008) Bayesian regime-switching model. We surmised that poor economic policy, both monetary policy and regulatory policies in the 1970s (think price controls, lines at gas stations, economic malaise, etc.), caused the spillover from energy to productivity. But more enlightened approaches since then have broken this correlation.

So any dumb economist (or AI model) that blindly feeds 80 years of data series into a forecast model without recognizing that there have been potential structural shifts in how energy shocks propagate through the economy will likely create false alarms looking at today’s energy shocks. I suspect Moody’s told some interns to come up with a catchy forecast. So they looked for all economic and financial series that correlated with the business cycle in the past, then screened for those that look really poor today (oil prices, likely equity volatility in March and early April, a yield curve inversion a few years back, etc.) and ignored the ones that still look solid or even strong today. And voila, you have a forecast model “That Has Never Been Wrong” and now strongly predicts a recession. It’s likely economic junk science and data snooping. Though if they do this regularly, they will certainly be right eventually, and then pat themselves on the back.

More reliable recession indicators than an oil price shock

If oil shocks are no longer reliable economic indicators, where else should we look? Here are my favorite indicators. You may recognize the first three, because I’ve introduced them previously as my preferred financial/economic indicators for pretty much as long as I’ve been blogging about economics. The earliest blog post on my blog discussing those three indicators was Market Timing and Risk Management, Part 1 – Macroeconomics, published in February 2018. The reason I mention that is that I don’t want folks to accuse me of cherry-picking indicators that fit my current narrative, which is what I believe Moody’s is doing. Rather, the indicators are the ones I’ve used both privately and professionally for as long as I’ve run economic forecasting models.

1: Weekly Unemployment Claims

I think I mentioned this multiple times before; if I could only monitor one single economic data series, this would be it: weekly unemployment claims. It’s a reliable economic indicator of labor market health and has the advantage of not being revised (unlike the dreaded monthly Payroll Employment numbers). Every historical recession saw a rise in unemployment claims around the turning point, sometimes even well before. But the current readings are very solid in the low-200,000s. Unless we see this number spike up to at least 300,000, I wouldn’t bet on any recession starting anytime soon. The recession peak in the claims series should be at about 400,000 or above. I would give this indicator an A-.

2: The PMI Index

PMI stands for the Purchasing Managers Index. You can find the dataseries here. Values below 50 indicate contraction. I consider readings between 45 and 50 a mere warning sign, though. Normally, the PMI needs to fall below 45 to signal a recession; please see the chart below.

The most recent reading is the 3/31/2026 value of 52.7, published on April 1. Note that this figure already includes data collected during March 2026, when the WTI crude price was hovering around 100. This fact hasn’t caused any alarm bells in the manufacturing sector yet. In fact, over the past few years, the PMI has been in that 45-50 limbo state, where I had my worries about the index slipping below 45, which historically has had 100% recession predictive power. But luckily, we’re now well above the danger zone. Of course, this can all change rapidly, but right now, this indicator looks like a B-.

3: The slope of the Yield Curve

Maybe the reason some economists “hope” for a downturn around the corner is that one of the historically most accurate recession indicators turned red between 2022 and 2024, yet there was no subsequent recession. Only a slowdown and a bear market in 2022. True, the indicator has previously led the business cycle, i.e., the yield curve inverted and caused a negative 10y-2y spread, sometimes a year or more before the turning point. Indeed, that was always the selling point, because very few indicators lead the business cycle by that much. But the recent yield curve inversion first took hold in April 2022 (briefly, for only two days) and then remained inverted between July 2022 and August 2024. But it would be a stretch to call a 2022 yield curve inversion a predictor of a 2026 (or even early 2027!) recession. If you’re in an expansion in month M, the unconditional probability of a recession between months M+1 and M+48 is almost 64%. So, if I used my super-secret signal, “it rained today in the Pacific Northwest,” as a recession indicator, it would have an accuracy of almost two-thirds over the next four years. Maybe good enough for the interns at Moody’s, but how useful is this really?

So, this indicator gets a B-. The current reading is certainly very, very solid, and it looks like there is no immediate risk. My theory is that the 2022 yield curve inversion correctly predicted the 2022 economic slowdown and bear market. Thanks to the still-generous stimulus programs, the slowdown never turned into a full-blown recession. However, we shouldn’t “recycle” the 2022 yield curve inversion to predict another recession now.

But some doubt still lingers: One concern is that an oil price shock will force the FOMC to raise interest rates again. I could also spin a story about how the rate hikes have done some damage to the plumbing in the financial system that we haven’t quite seen yet. Maybe all that private equity and private credit junk being pushed on retail investors is a warning sign that smart institutional investors are leaving the space and looking for others to hold the bag before that sector implodes. But there doesn’t seem to be much stress in financial markets, at least not yet, which brings me to the next two indicators…

4: High-Yield Spreads

Another reliable leading economic indicator that has, so far, barely moved is the spread that risky (high-yield) borrowers need to pay over the least risky fixed-income instruments (e.g., Treasurys). In legitimate recessions, this spread rose even prior to the official start of the NBER-declared business cycle peak. It’s a highly useful leading indicator. An exception would be the pandemic, where the yield spread was “only” a coincident indicator, for obvious reasons. Right now, the yield spread sits at only 2.95 (as of 4/13/2026), close to its all-time low and well below where I’d worry about a recession. In fact, at this point, we haven’t even reached the yield spreads during historical garden-variety hickups and other non-recessions and false alarms, like 2018 and 2022, let alone the nothing-burger tariff tantrum of 2025. During a recession, we’d need to see double-digit spreads, and leading up to the event, we’d need a 5.0-7.5 reading in this indicator. None of this has materialized yet. This indicator gets an A-.

5: The S&P500 index. Duh!

If the oil price shock was such a big threat to the economy, why hasn’t that already shown up in, well, the most scrutinized, analyzed, and heavily traded financial betting market, i.e., the U.S. stock market? There’s almost $70 trillion of U.S. total market capitalization. That’s a lot of money, and thinking this is a massive blob of dumb money that hasn’t caught up to the smart, enlightened models at Moody’s and the Motley Fool is a bit ludicrous. The more likely scenario is that the market has shrugged off the energy price shock. We haven’t even seen a correction, much less a bear market, yet. Quite the opposite: the S&P 500 price index is within a few points of its all-time high, and the Total Return index just snatched a fresh all-time high (as of the 4/14/2026 close). The stock market certainly suggests we will not enter a recession anytime soon. In almost all previous recessions, we’d see marked drawdowns in the S&P well before the recession. This indicator receives an A (as of 4/14/2026).

Conclusion

To wrap things up, do I believe the U.S. economy is in imminent danger of a recession? Likely not, at least not from the oil price shock alone. Most other indicators look either solid or strong. That doesn’t mean there is no chance of a recession. As we learned above, there is a roughly 20% probability of slipping into a recession, unconditionally on any additional information. Maybe, with the uncertainty about how the Middle East tensions will work out, I’d bump that probability up by a few percentage points. But it’s far too early to call a recession at this point. No need to panic! Happy investing and happy retirement planning, everyone!

Thanks for stopping by today. I am looking forward to your comments and suggestions below.

Title picture credit: WordPress AI