Xerox has announced a delay in its auction of 239 patents, planned for July 24-29, citing “numerous requests” from potential bidders for additional time. The company also increased the number of patents to 546. While a skeptic may claim the delay is really due to a lack of interest, I think it’s more likely that there’s a great deal of genuine interest – but the interested parties are struggling to determine exactly how much they should bid. Patent valuation is always part science, part (prior) art, and never easy.

According to numerous reports, these Xerox patents, which “came out of research work at its various global R&D centres”, are to be auctioned in lots valued at between $1 million and $10 million each. The auction web site suggests initial bids for each lot, ranging from “low six figures” to “low seven figures”. There are 27 lots covering technology fields such as audio communication, touchscreens, image compression, networking, and print systems. This means that Xerox is hoping to rake in between $27 million and $270 million. Even at the low end of the range, that’s not chicken feed.

The bidders have to weigh the up-front purchase cost ($27 million plus, in this example) against what they reasonably believe they can obtain in licensing fees from operating companies that are infringing the patents – and that do not already have a license from Xerox.

Naturally, it’s essential for the bidders to scrutinize the patents to confirm that they are well drafted, and not likely to be overturned by an invalidity challenge. Furthermore, they’ll want to look very carefully for evidence of use of these patents in the products of prospective licensees. A diligent licensing company does not want to skimp on either step, regardless of the seller’s timetable.

Now, the corollary is that this analysis can end quickly if the patents are not of high quality. Mark Peterson, CEO of Robinwood Consulting, has expressed doubt that the patents are particularly good. But I believe this delay proves the opposite. A rigorous valuation can take weeks or months, depending on the breadth of the portfolio. Xerox’s original schedule gave less than four weeks. The extension to September 11-16 is much more realistic; it could even be extended again – we’ll see.

Selling excess patents is good business

What’s notable about patent sales from major companies is the huge range of values, such as Kodak’s (at a disappointing $94 million) and Nortel’s (at an eye-popping $4.5 billion). But unlike Kodak and Nortel, Xerox is not bankrupt. Why would a profitable company like Xerox be selling its patents? First of all, these 546 patents represent only 4% of the approximately 12,000 patents owned by Xerox.   This sale is better understood as an act of responsible asset management than one of desperation.

Senior management at many public companies are learning that their fiduciary responsibility includes proper use of patent assets. This is leading companies to look for licensing opportunities – by starting their own licensing programs or in partnerships with companies such as Conversant – or the outright sale of parts of their patent portfolio.

In fact, Xerox has sold patents in the past, although this is their first auction. They will doubtless continue to do so in the future, just as its engineers will continue to file patents on new inventions – currently at a rate of about 1,000 per year. A functioning secondary market for patents, and a strong legal framework for their enforcement, ensures that companies will be able to generate additional returns on their R&D investment, even for inventions which they do not themselves practice.