Policy Considerations: The On-Sale Bar for Secret Processes

by Dennis Crouch

In light of the Federal Circuit’s recent decision in Celanese v. ITC, it’s worth examining the policy implications of maintaining a strong on-sale bar that extends even to invalidate patents on secret processes when the resulting products have been commercialized. This rule is rooted in pre-AIA jurisprudence and is now affirmed under the AIA.

1. Early Public Disclosure vs. Trade Secret Retention

One of the primary arguments in favor of a strong on-sale bar is its potential to encourage early disclosure of inventions. The theory is that by forcing inventors to choose between patent protection and trade secret protection early on, we promote the progress of science and useful arts by making knowledge publicly available sooner.

However, this assumption may not always hold true. In some cases, particularly where the process is difficult to reverse engineer from the product, this rule might instead incentivize inventors to maintain their processes as trade secrets indefinitely. This could result in a net reduction of public knowledge, contrary to the patent system’s goals.

The actual impact likely varies by industry and technology type. For instance, in fields where rapid innovation is crucial and patent protection is highly valued, the on-sale bar might indeed spur earlier disclosures. In other sectors where trade secrets are traditionally favored, it might have the opposite effect. Empirical research on this topic could provide valuable insights. Studies comparing disclosure rates and innovation patterns in jurisdictions with and without such an on-sale bar could help policymakers assess the rule’s real-world impact on public knowledge dissemination.

We also need to recognize that the AIA’s move to a first-to-file system also provides added incentive for prompt filing, thus reducing the need for the on sale doctrine to serve this purpose.

2. Preventing Unfair Patent Term Extension

A key rationale for the on-sale bar is to prevent de facto patent term extension. Without this rule, inventors could theoretically exploit their invention commercially as a trade secret for years before seeking patent protection, effectively extending their monopoly beyond the statutory patent term. This principle was articulated by Judge Learned Hand in his seminal Metallizing Engineering decision:

Metallizing Engr. Co. v. Kenyon Bearing & Auto Parts Co., 153 F.2d 516 (2d Cir. 1946). Judge Hand’s decision underscores the fundamental bargain at the heart of the patent system: exclusive rights in exchange for prompt public disclosure. This rationale continues to influence modern interpretations of the on-sale bar, including its application to secret processes. Of course, even the great Learned Hand struggled with this issue. He had originally decided in this type of case that the inventor did not lose his rights, but was subsequently convinced in Metallizing Engineering to change his mind — acting without the en banc authority that would be required today.

It is worth noting that there could be alternative solutions to this problem that don’t result in complete invalidation of the patent. For instance, Congress could consider a system where the patent term is reduced based on the duration of pre-filing commercialization. This would offer a different balance of inventor’s interests against the public’s interest in timely disclosure. Such alternatives, however, would require legislative action and are beyond the courts’ authority to implement under the current statutory framework.

3. Fairness to Competitors and Prior User Rights

The on-sale bar also promotes fairness to competitors by preventing situations where they might be sued for infringement years after a product has been on the market, when they had no way of knowing about the secret process. However, it’s important to note that Congress has already addressed this concern through the prior user rights provided in 35 U.S.C. § 273. This provision offers a defense to infringement for those who commercially used the invention at least one year before the patent’s filing date.

That said, § 273 has been criticized for its numerous caveats and limitations. For instance, it doesn’t apply to patents owned by universities or their technology transfer organizations, and it’s limited to the specific site where the prior use occurred. These limitations may render § 273 less effective in practice than it appears on paper.

4. Avoiding “Traps for the Unwary”

Any rule that results in either a windfall or complete loss of rights should ideally be simple and predictable, allowing interested parties to plan accordingly. This is especially crucial for small entities and individual inventors who may not have ready access to sophisticated legal counsel to plan a complete legal strategy. The current on-sale bar, particularly as applied to secret processes, could be seen as one of those “traps for the unwary.” Such “traps” can create pessimism and uncertainty that may undermine the patent system’s incentive structure.

5. International Harmonization

Finally, we must consider the goal of international harmonization in patent law. The AIA was partly motivated by a desire to bring U.S. patent law more in line with international norms. However, the on-sale bar as applied to secret processes appears to diverge from the approach taken in many other jurisdictions.

For instance, the European Patent Convention does not have an equivalent to the U.S. on-sale bar. Under European law, selling a product made by a secret process generally would not bar a later patent on that process, provided the process cannot be deduced from the product itself.

This divergence raises questions about the necessity of maintaining this particular aspect of U.S. patent law. While there may be valid policy reasons for the U.S. approach, the benefits of international harmonization – including reduced complexity for inventors seeking global protection and increased predictability in international trade – should be weighed against these reasons.

What are your thoughts? Should non-enabling commercialization bar later patenting?