Editor's note: This is the second post of a two-part series on BNPL.

A friend of mine owns a popular cheese shop, and for the longest time, he resisted upgrading to checkout scanners. Chatty cheese experts painstakingly keyed in prices on an old-style cash register while customers gazed at the mouthwatering array of cheeses, olives, figs, pâté, and crackers from around the globe.

"If it takes longer to check out, people buy more," my friend said.

My friend's theory gives some hints into the merchant business model for buy now, pay later (BNPL). The corollary would be something like, if the customer can take longer to pay, the customer will buy more. The lure for the merchant is that customers will buy more and shop more often and that new patrons will come through the real or virtual doors when they can enjoy now and pay later.

Last week, my colleague Dave Lott wrote about "stormy waters" for consumers and the firms offering BNPL services. It's not all smooth sailing for merchants, either. Buy now, pay later is expensive for merchants, with a price tag as much as 7 percent of the purchase priceOff-site link, compared to 3 percent or less for credit or debit card payments. While BNPL shifts the risk of nonpayment from the merchant to the BNPL provider, offering BNPL to the customer is much, much more expensive for the merchant than accepting a cash payment.

In addition, BNPL could be problematic for customer satisfaction in the long run. Remember how it feels when you take a vacation in February and then the credit card bill comes due in April? When we consume something before we pay for it, we can be less satisfiedOff-site link with the product or service and what it cost. In addition, customer service can suffer when merchandise is returned but it takes time for the BNPL vendor to cancel the scheduled payments and refund any amounts already paid.

Also important, will BNPL continue to be effective at reducing shopping cart abandonmentOff-site link and increasing ticket size? What began as a way to recruit new customers and generate more revenue could devolve into yet another expense, a standard cost of doing business as more and more merchants adopt the practice. Like other merchant expenses, the cost of a no-interest-loan is getting rolled up into what I pay for a pair of cute shoes. Do I need the shoes? If I and other consumers decide to take a pass on such discretionary purchases, will the cost of BNPL to merchants still be worth the results?

This summer, with consumers fretting about the price of gas and groceries and the trajectory of their retirement investments, could BNPL get too expensive for merchants?