Coffee Bean Prices Are Falling Fast — Here's What a Cart Owner Should Actually Do About It

Arabica futures fell from $3.50 to under $2.80/lb. A real LA cart owner explains what it means for your menu and margins.

Joan Merrill On

Joan Merrill On runs a mobile espresso and matcha cart in Los Angeles, serving weddings, corporate events, and pop-ups.

I've been running a mobile espresso cart in LA for a while now, and coffee bean prices have been the line item that keeps me up at night. A 5-pound bag of decent specialty beans can swing 20, 30, 40 dollars in a bad month. So when I saw the news this week, I actually read past the headline.

Here's the short version: arabica futures fell to about $2.78–2.82 a pound in mid-September, the lowest in roughly ten weeks. Back in July, they were trading above $3.50. That's roughly a 20% drop in two months. The International Coffee Organization just forecast record global production for the 2025/26 crop year and said the market will be oversupplied. Brazil shipped 2.87 million bags of arabica in August — up nearly 26% from last year — and the first two weeks of September ran more than 50% ahead of last year's pace.

In plain English: the bean supply faucet turned on, and prices are responding.

Now, here's what I'm telling myself, and what I'd tell any cart owner: don't confuse the futures market with your actual costs, and don't touch your menu prices.

Your beans are not priced off the futures ticker

The futures price is the commodity price — the wholesale floor for generic arabica. Nobody I know buys that coffee. We buy from roasters, and roasters price off contracts, logistics, labor, and their own margins, not the September contract ticking around on a screen.

So the first question isn't "prices are down 20%, when do I get cheaper beans?" It's "when does this show up in my roaster's invoices?" The honest answer: usually 4 to 12 weeks later, and only partly. A roaster who raised your price when futures spiked isn't necessarily going to volunteer a cut when they fall. You have to ask.

So call your roaster. That's step one. Not next month — now. Ask them: is my price coming down, and when? If you have a relationship, they'll usually move if futures stay low for a few weeks. If you're buying through a distributor or a big national roaster with fixed pricing, you have less leverage, but the call still costs you nothing.

Do the math on your cost per drink — for real this time

Let me show you mine, because most operators I talk to have never actually run this number and they're leaving margin on the table in both directions.

Say you pay $18 for a 5-pound bag of espresso beans. That's 80 ounces. A double shot is 18 grams, call it 0.63 ounces. So you get roughly 127 drinks per bag. $18 divided by 127 is about 14 cents per drink in bean cost.

Wait, that can't be right, can it? Fourteen cents for the actual coffee in a $6 latte? Yeah, it can. Beans are not your cost driver. Milk is. Cups are. Your labor, your vehicle, your event fee — those dwarf the beans.

This is why I say don't cut your menu prices. If bean costs fall 20%, your cost per drink falls by maybe three cents. Customers won't notice a three-cent difference, and you'd be giving away margin for nothing. Falling bean prices are not a reason to lower prices — they're a reason your margins quietly recover from the beating they took when beans were spiking.

What I actually did this week

Three things:

1. I called my roaster. Asked when I'd see relief on my invoice. Turns out my price is set quarterly, so my next adjustment hits in October. Fine — now I know, and I've got it in writing.
2. I ran a scenario on my catering minimums. My Essential package is $895 for 75 guests over 2 hours. My variable costs per guest — beans, milk, cups, syrups — run me about $2.10 per drink. If beans fall another 15% and my roaster passes even half of it through, my per-guest cost drops maybe 5 cents. Tiny. So I'm not changing my packages. But if your minimums were priced during the panic earlier this year, it might be worth re-checking whether you built a bean-cost cushion into your pricing that you no longer need — because that cushion is now pure margin, and knowing it's there lets you be more flexible on discounting to win bids.
3. I considered locking in a contract. Some roasters will let you forward-buy beans at today's price for the next 3–6 months. With futures at a 10-week low and a record harvest coming in, a lock-in is a bet that prices stay low or rise again. I'm not doing it yet — one weird weather event in Brazil and the bet flips — but if you burn through 40+ pounds a week, it's worth asking your roaster what a forward contract looks like. Volume buyers get real options here.

The one caution

There's a weird wrinkle in the data worth knowing: ICE certified arabica stocks are at 217,646 bags, the lowest since 1999. That's the exchange's deliverable supply — the physical coffee that can actually be delivered against futures contracts. So futures are falling because Brazilian supply is flooding export channels, but the exchange warehouse cushion is historically thin. If that Brazilian flow slows down or a weather scare hits, prices can snap back fast.

Translation: enjoy the relief, don't bank on it permanently. Keep your pricing as-is, keep your margins, and don't sign anything long-term at a discount unless the math actually works for your volume.

Bean prices have been a gut punch for the last couple of years. This pullback is real and it's meaningful. But the cart owners who benefit most aren't the ones who react to headlines — they're the ones who know their cost per drink cold and negotiate their roaster contract like a line of business, not a habit.

MorningStorm.co is my booking platform for coffee cart owners — storefront bookings, invoicing, and QR ordering built for how carts actually work. The beta is free and open to 10 cart owners. Check it out at https://morningstorm.co