Based on incoming economic data, it would appear as if it’s the labor market that’s holding the economy together. But you have to wonder, just how long can it last?

According to the Federal Open Market Committee (FOMC), the answer is “a while.” In its latest projections, the FOMC penciled in an unemployment range of 4.4%–4.7% for 2023, not much higher than the current 3.5% rate, followed by a 4.3%–4.8% range in 2024 and an improvement to 4.0%–4.7% in 2025. Goldman’s chief economist, Jan Hatzius, expects “a trough to peak increase in the unemployment rate of 0.7 percentage points (pp), roughly one-third the increase seen in even the shallowest US recessions.” The Conference Board is on the same page as the Fed, forecasting a 4.5% rate in 2023. Even one of the more pessimistic economic teams on the Street, BofA, isn’t particularly grizzly. It is looking for 5.5% unemployment this year.

Here are a few examples of what could go wrong…

A net 31.3% of banks that participated in the New York Fed’s Q4 Senior Loan Officer Survey said they were tightening lending standards on their entire loan book, the fifth time in as many quarters that the series has deteriorated. A reading like the current one also rolled into town in Q4/2000 amid stock market declines. At the time, the unemployment rate was doing just fine, hanging out below 4%. But it didn’t last long: unemployment rose to 6.1% a couple of years later.

Similarly, the survey hit a response rate that was this ugly in the summer of 2007, when unemployment was 4.7%. The rate doubled from there, but then again, not many people were anticipating a collapse of the nation’s banking system at the time. Because of Dodd-Frank Wall Street Reform and Consumer Protection Act, it’s unlikely that 2023 or 2024 will confront something similar.

Nevertheless, we don’t think you need a total systemic come-apart to turn a jaundiced eye toward the FOMC’s sub-5% unemployment prognostications over the next 35 months.

After all, housing is ice cold. Sure, 100% loan-to-value mortgages are a distant memory from the halcyon days of 2004–2005. But today’s situation has its own brand of peculiarity: many people are stuck in their current place, current on their 3% mortgage but unable to pack up if it means doubling their mortgage rate elsewhere. A dearth of activity cannot be good news for roofers, masons, pest control teams and anyone else whose job heavily relies on inspectors pointing out homes’ flaws.

The National Association of Home Builders (NAHB) survey’s projected new home sales gauge has fallen harder this cycle than any other time on record, second only to the last generation’s housing debacle—and only by a hair at that. Whether the NAHB data leads employment conditions by a year or whether it’s more like two years is up for debate—but it did start waning 13 reports ago. For now, we will all have to wait to see if the labor market cares.

Meantime, it’s hard to get our minds off the -32.9 print on the January Empire State Manufacturing Survey. Draw a line through its -78.2 bomb during COVID-19’s first inning, and you land on the worst of it coming in February 2009, when it was -34.3.

Still, February 2009 should grab your attention anytime someone mentions that month; things had gotten so bad by then that stocks decided there was nobody left to sell anymore. The crisis bear market ended just days later, on March 9, 2009. From that lens, having such a miserable Empire State report could be ideal for contrarian equity longs. Then again, unemployment in February 2009 was 8.3%, not too far off what would ultimately be the worst of it, 10%. Today, you’ll catch gasps of disbelief if you hasten to whisper that maybe, just maybe, 2023 unemployment could breach 5%. While this is not our base-case scenario, such an outcome would fall under the heading “what could go wrong” this year.