API: You often get texts from computers; eventually they’ll be taxed

I pulled out my phone just now and checked my latest text messages. Of the 12 most recent texts I received, four were from automated sources: our local transit agency, warning there could be delays in my neighborhood; a contractor we’d hired for a project, sending me an automated message that he was on his way from his last job; a local restaurant, confirming that I’d placed a takeout order that would be ready in 15 minutes; and our credit union, confirming that I had successfully completed my password update.

Other texts I received recently include one from a food delivery service, offering me a discount coupon if I place an order using its app this weekend, and one from a ticket broker, letting me know tickets are available for a number of upcoming local events.

No humans were directly involved in the transmission of these messages. Everything was automated. For example, I signed up for the transit system messages because I sometimes take the bus for events. (Why pay $30 for parking near the stadium and deal with game-day traffic when I can leave my car at home for free?) Now I get automatic text alerts when there are schedule or service changes on my routes.

Similarly, the food delivery app knows my family often orders takeout for dinner on Fridays, so the app designers built in a weekly nudge to encourage us to do it again. Likewise, the ticket app knows what kind of events I’ve bought tickets for in the past, so it encourages me to buy more tickets when similar events are coming up.

This kind of communication uses an application programming interface, or API. APIs allow two different pieces of software to share information. The app on the contractor’s phone and the text-messaging software on my phone, for example: In this case, the data it shared was: “I will be there in 20 minutes.”

There are a large number of APIs for internal business use as well. API communications is one of the technologies that enables the Internet of Things.

Automated vending machines, for example, can use APIs to alert their company’s home offices when the cash box is full or the Snickers supply is empty. APIs can be used to monitor remote equipment and alert maintenance teams when that equipment needs attention.

Tax preparers can even get an API to share data with the Internal Revenue Service about their clients’ tax returns.

Likewise, the ever-more-common smart appliances (smart refrigerators, robot vacuums, smart ovens) also use APIs to communicate their status to users in the home.

API-based communications aren’t currently taxed

API-based communications can be tricky to classify for tax purposes. To understand why, we need to look back on where communications taxes began, decades ago — with telephones.

Telephone calls were originally placed by one party to another and carried as electronic signals over a network of copper wires and poles that typically were located on public rights-of-way. Governments taxed those communications — often with fees for the use of the rights-of-way — and wrote tax laws that applied to telephone calls and similar services. Those taxes continued to apply even as the technology changed, wireless services began replacing some of the copper-wire connections, and text messages replaced many of our voice calls.

But APIs aren’t necessarily the same. With an API, by definition, what you have is two software programs sharing data. That can be your car sending your phone a message that it’s time for an oil change, your smart refrigerator sending a message to your tablet that you’re out of milk and eggs, or your local burger joint sending a message to your smartphone that dinner almost is ready. Those messages are machines talking to machines, even if the point of the machines talking to each other is to alert us humans of something we need to know.

And yet, the API-based message from the contractor last week looks just like the text message my brother-in-law had sent me a few days before to let me know he was on his way over. The only difference is that he typed the text message with his fingers, while the contractor’s message came from an automated data-sharing app.

As the line between the two types of messages continues to blur, the definitions of these services for tax purposes becomes ever more important.

States could provide clarity

Most states already have laws on the books addressing the taxability (or nontaxability) for sales of wireless phone calls, texts, or SMS messages. However, the intent of those laws was likely aimed at taxing those items when they are billed by a cell phone carrier like Verizon or T-Mobile. Does the same logic apply when a contractor management app bills users for API communications sent from a platform in the cloud? Some of those taxes are clearly aimed at phone carriers, so does an app or cloud platform fit under that definition? Reasonable minds might come to different conclusions.

Unfortunately, this leaves us in a familiar place in the communications tax technology space: waiting for tax jurisdictions to provide clarity or precedent around their position. Once that happens, we should start to see movement on this issue, hopefully with some definitions that will help determine which of the automated communications now pouring into your devices should be taxed, and which should not. Keep checking back for updates.

In the meantime, if you need help figuring out communications taxes in their current state, check out our Communications Tax Resource Center.

Sales tax rates, rules, and regulations change frequently. Although we hope you'll find this information helpful, this blog is for informational purposes only and does not provide legal or tax advice.