If you run an open-source project that pays contributors in tokens, you need to understand the Howey Test. Not because you are a securities lawyer. Because the SEC does not care whether you think your token is a utility -- they care whether it walks like a security, swims like a security, and quacks like a security. This article is a developer's field guide. We will walk through the legal framework, then apply it to real patterns you see in bounty token projects. What Is the Howey Test? In 1946, the U.S. Supreme Court decided SEC v. W.J. Howey Co., a case about Florida orange groves. The Howey company sold tracts of citrus land along with a service contract to cultivate and harvest the fruit. The buyers did no farming. They just collected checks. The Court ruled this was an "investment contract" -- a security -- because it met four conditions. These four prongs are now the standard test for whether any asset is a security under U.S. law: An investment of money -- Someone pays value to acquire the asset. In a common enterprise -- The investors' fortunes are pooled or tied to the same venture. With an expectation of profits -- The buyer anticipates returns. Derived primarily from the efforts of others -- Those returns depend on work done by a promoter or third party, not the buyer. All four prongs must be met. If any one fails, the asset is not a security under Howey. This sounds simple. It is not. Let us look at how it plays out for bounty tokens. When Bounty Tokens Are Likely Securities Consider a hypothetical project: CoinBounty. The founder mints a token on a smart-contract platform, sets up a bonding curve for public purchase, and announces: "Earn tokens by contributing to our repos! Also, buy them on our bonding curve -- early buyers get the best price." Let us apply Howey. Prong 1: Investment of Money The bonding curve is a purchase mechanism. Users send SOL, ETH, or stablecoins and receive CoinBounty tokens in return. That is an investment of money, full stop. It does not matter that some tokens are also earned through work. If there is a purchase path, Prong 1 is met for every token acquired through it. Prong 2: Common Enterprise All token holders share a common pool. When the founder markets the token, everyone's holdings rise. When interest fades, everyone's holdings fall. The fortunes of buyers and contributors are tied to the same venture. Prong 2 is almost always met for fungible tokens. Prong 3: Expectation of Profits If the project's Discord says "get in early," "token will moon," or "we're listing on [exchange] next month" -- that is an explicit expectation of profits. Even without those statements, a bonding curve by construction implies that early buyers profit from later buyers. The mechanism itself creates the expectation. Prong 3 is met. Prong 4: Efforts of Others This is the killer prong for most bounty tokens. Ask: where does the token's value come from? If the answer is "the founder's marketing, the founder's exchange listings, the founder's partnership announcements" -- that is the efforts of others. The buyer sitting on a bonding curve position is not doing anything. They are waiting for the founder to make their tokens worth more. All four prongs met. CoinBounty tokens purchased on the bonding curve are likely securities. Even the earned tokens become legally complicated when there is a parallel purchase market, because the existence of a liquid speculative market changes the "expectation of profits" analysis for everyone. When Bounty Tokens Are Likely NOT Securities Now consider a different model. A project has been running for over a year. There is no token sale. No bonding curve. No exchange listing. The only way to acquire the token is to do real, verifiable work. Prong 1: Investment of Money No one purchases the token. Contributors earn it by writing code, mining on real hardware, running infrastructure, or completing audits. There is no investment of money because there is no purchase mechanism. Some legal scholars argue that contributing labor constitutes an "investment." Courts have generally rejected this when the labor produces standalone value (code that works, infrastructure that runs) rather than being purely speculative (buying a lottery ticket). Prong 2: Common Enterprise This prong often still applies -- contributors and the network share a common interest. But without Prong 1, the analysis is already weakened. Prong 3: Expectation of Profits If the token has real utility -- paying for network fees, purchasing compute jobs, settling agent-to-agent transactions -- then holders use the token, they do not just hold it hoping for appreciation. The token is more like arcade tokens at a bowling alley than shares in a company. Consider a concrete example: a token earned by attesting real hardware (PowerPC G4s, SPARC workstations,