About five minutes after Illumina announced their intention to acquire PacBio, my phone started ringing with market analysts, fund managers, and journalists all looking for the answer to one question: Will this merger pass anti-trust review? While I can’t say whether it WILL, I can emphatically say that it SHOULD. Read on to see why I think so.
The major concern, of course, is that the dominant DNA sequencing company is looking to acquire a smaller competitor, further consolidating their stranglehold on the market. Not only that, but a competitor which claims to have a superior technology that, with imminent developments, will be cost competitive. That’s why both the FTC and the UK’s CMA are looking at this deal very closely. The main question they seem to be asking is do Illumina and PacBio address the same market. This is kind of a hard question to answer as it’s a moving target, but right now the answer is mostly “no” – Illumina’s short reads and PacBio’s long reads are generally used for different purposes. But that doesn’t mean there isn’t customer overlap and that there aren’t some blurry lines (which will get blurrier as PacBio continues to improve).
Instead of asking if Illumina and PacBio markets overlap, I like to think about objections that customers and competitors might be making and what the market would look like if this deal doesn’t go through. Let’s deal with customers first. Their three main concerns would be would be what happens to the price they have to pay, what kind of support they’ll get, and if the PacBio technology will still be available if the deal goes through.
Pricing: Current customers of PacBio and Illumina might be worried about prices. If PacBio was about to start putting pricing pressure on Illumina (something they haven’t experienced in years), this would let Illumina prevent that from happening. But that isn’t a valid concern because we’re already pricing pressure from other sources – Oxford Nanopore (ONT) and China’s BGI (and subsidiary MGI). With the latest advancements in ONT’s PromethION platform, they’ve already reached the 30X $1000 genome. Now, it’s hard to make a straight comparison with Illumina’s sub $1000 genome for two reasons. First, ONT generates really long reads, so their genome will be substantially more complete with fewer gaps – point to ONT. Second, while ONT’s error rates have improved, they’re still substantially higher than Illumina’s – point to Illumina. BGI is a much more “apples-to-apples” comparison as it’s also a short read SBS-based system. BGI hasn’t launched their platform globally yet, but they clearly have plans to do so. And with an advertised rate of $600 per 30X human genome, any commercial success they have will start putting pricing pressure on Illumina. It’s really just a question of how likely customers will switch over to BGI, a company that not everyone trusts.
Support: A potentially more valid concern of PacBio’s current customers is losing the wonderful technical support from PacBio they’ve come to rely on. While Illumina also has a strong support team, the company is starting to get a reputation of being arrogant and not taking customer complaints too seriously – a common trend in companies who become so dominant. But a company no longer being quite so friendly or collaborating so closely with their customers isn’t the same as being anti-competitive. The PacBio customers might just have to get used to what everyone else has been putting up with for years.
Availability: The other potential concern I’ve heard analysts ask about is the future availability of the PacBio platform – might Illumina just mothball the platform once the acquisition is done, getting rid of the competition once and for all? Sure, they could, but that would be a very expensive way of getting rid a competitor without addressing the actual threat of price competition (ONT and BGI). Oxford Nanopore might be happy if Illumina did this, but it wouldn’t help Illumina at all.
So how about what competitors think? They really only have one thing to complain about – that the acquisition would make Illumina too dominant. But who is there to complain? Oxford Nanopore’s CTO Clive Brown has already publicly stated that PacBio isn’t a threat.
Maybe Thermo Fisher could complain, but they really haven’t been doing much to advance their Ion Torrent platform. If competition with Illumina (or PacBio) were really a concern, they should have been doing more the past several years.
The other way to think about this is to ask what would happen if the deal doesn’t go through? Sequencing prices will be lower with or without the acquisition, so the real question is what happens to PacBio? The hard truth is that while PacBio’s Sequel platform would be considered a big success by pretty much every measure, the company itself is still not a commercial success. After many, many years they are still not profitable. That can only go of for so long before investors stop buying into the dream. So that means they’ll either go out of business or get acquired by someone else. But who? Roche? [insert your own 454 and Genia joke] Thermo? They aren’t doing so hot with Ion Torrent. Oxford Nanopore? They don’t seem to think too much of it. Maybe some other large company who wants to buy their way into the sequencing space, like GE or Agilent, but I don’t think there’s a convincing argument that they would do a better job at keeping PacBio around than Illumina would. Maybe BGI would try like they did with Complete Genomics, but I’m not sure if CFIUS would go for it this time around.
One thing regulators sometimes do is force the acquiring company to divest part of itself to prevent it from being so dominant. But would could Illumina divest? They could spin off the microarray business, but it wouldn’t have much of an impact on the sequencing market. They could spin off the short-read technology, but that just means they would become PacBio – they’re not going to do that. So it seems like letting this deal go through “as is” is the best thing for the market. For the sake of sequencing customers everywhere, let’s hope the regulators see things the same way I do.
