Fed Chair Jerome Powell yesterday admitted to Congress that inflation effects in the U.S. economy “have been larger than we expected, and they may turn out to be more persistent than we have expected.” The U.S. Consumer Price Index rose 5% year-over-year in May, the highest level in 13 years. Inflation excluding food and energy was at 3.8%, the highest level for that number in 30 years.

That’s all worrying, and inflation has rightly been a focus of discussion as the pandemic ends and stimulus checks keep going out. Continued 5% inflation or higher would be, among other things, a vindication of the anti-Fed, hard-money thesis that fuels many Bitcoiners, and a real chance to test whether the digital currency is an anti-inflationary hedge.

But Powell didn’t sound any major alarms on Tuesday, holding to the line that inflation effects in the economy were largely transitory, the product of post-pandemic reopening demand. Powell cited pandemic demand for airline tickets, for instance, as everyone gets vaccinated and decides to travel at the same time.

He might also have cited the price of rental cars, or used vehicles, or graphics cards, or Uber rides, or housing-grade lumber. There are certainly other examples, but these may be the most representative of what you could roughly call supply-driven inflation. In short, rising overall prices in the U.S. have been substantially driven, not by a surplus of money above pre-pandemic levels, but in many cases by supplies of goods and services that are below pre-pandemic levels.

Here’s a picture I took of a used car lot in Texas recently.

Your mental model for inflation may be someone taking a wheelbarrow full of money to the grocery store to buy a loaf of bread, but maybe add this image, too – a near-empty used car lot in a nation full of people going absolutely insane with the desire to move.

The complexity of the used-car story is representative of the broader supply-constraint story. Used car prices are surging because many car manufacturers cancelled their orders for crucial chips in early 2020, betting the pandemic meant car sales would crater long-term. They were incredibly wrong, but by the time they decided they needed more chips, fabricators had taken on other, more valuable clients. So now, the OEMs can’t make as many new cars, and people are turning to used vehicles en masse.

The chips that displaced those automaker contracts probably sometimes included graphics cards, or GPUs, which have also surged in price for reasons unrelated to what we consider “inflation.” The skyrocketing price of Ethereum in particular during the pandemic juiced GPU demand from miners, leaving gamers absolutely forlorn. The prices of computer peripherals like GPUs are part of the CPI calculation (though CPI measures retail prices, not the huge markups for GPUs on eBay in recent months).

The other stories are similar. Lumber prices went bananas in April and May because the people wanted new houses at rates well above pre-pandemic levels. Recent months have seen more new housing starts than at any time since roughly 2005.

Some of this demand is enabled by pandemic relief, of course. But even that doesn’t make it inflation as commonly understood. The key issue here, as Nobel-prize winning progressive economist Paul Krugman laid out yesterday, is the difference between transitory inflation and core inflation. And many of the current big inflation drivers have clear solutions that don’t involve constraining the money supply.

Partly, we’re already seeing a dropoff in demand as those who needed their houses/plane tickets/new cars YESTERDAY get their needs met. Lumber is already cratering, down 5% today as I write this. Growth in used vehicle prices has slowed to about 0.6% per month so far in June, down from rollicking 4.65% month-over-month growth in May. That’s extremely significant because used car prices made up a large portion of the May inflation numbers, and they look like they’ll be effectively gone for June CPI calculations.

But other solutions will be on the supply side, as disrupted supply chains reconstitute themselves to meet demand. Carmakers will eventually get their chip suppliers back (especially if Ethereum keeps dumping), and those used-car lots will fill up again. Spread across sectors, that would mean the last few months weren’t about long-term inflation, but short-term recovery, as dollars chase scarce goods and motivate producers to make more goods. Inflation happens when a growing supply of money hits a fixed supply of products. It seems pretty clear that’s not what we’re dealing with here when it comes to manufactured goods.

(The story is significantly different and more complicated when it comes to labor and wages, and I’ll point you to Joe Wiesenthal for a deeper dive on that side).

We’ll have a pretty definitive answer to these questions within the next couple of weeks; June inflation data will be trickling in until we get updated CPI in the second week of July. My bet is that as the economy smooths out its last few pandemic wrinkles, inflation will retreat, if not to near-zero prepandemic levels, at least to the 2% to 3% that the Fed, and most economists, consider manageable.