I swore to myself I wasn't going to write any more about the business of photography until the CP+ show, which is about the point in the year where we have enough data to fully understand the past year.
Unfortunately, an article published in Nikkei with comments from Canon's president Fujio Mitarai is raising a firestorm among the photography community. The typical headline that's producing all the hysteria is "the camera market will shrink by 50% over the next two years."
The camera market is a many-pieced thing. At one time in the digital era, the bulk of the market was compact cameras. We also have action cameras and video cameras to consider as part of the "camera market."
We've been losing about 30% or more volume (and dollars) each year in the compact market for quite a while now. Nothing seems to be changing there. 2018 will likely show the same decline from 2017 as 2017 did from 2016. And I see nothing that's going to stop the same decline in 2019, either. But it's not Coolpix and Powershot and Cybershot users that are getting all riled up on the Internet about Canon's statement, it's the DSLR and mirrorless users (ILC).
Here are the trailing 12-month numbers for ILC volume out of Japan for the past eight years, in millions of units: 11.5, 11.4, 11.4, 11.3, 11.2, 11.0, 10.9, 10.8. A clear decline, but a gradual one. Mostly caused by DSLR sales going down. Fortunately, mirrorless sales have been relatively stable over that same period. What we're seeing is a slow transition from DSLR to mirrorless in ILC, with a gradual decline overall ILC volume.
Not all is well with that last statement. In my more specific data I see several things that don't bode strongly for ILC in the next few years: (1) a lot of the mirrorless purchasing was sampling by DSLR users trying to see if "mirrorless was there yet"; (2) there's significant hangover in the inventory chain; and (3) the camera companies have been using pricing aggressively, to the point where their margins on the lower-end products have eroded.
My personal prediction for ILC is that we're going to see another 10% decline in volume in the next two years. Dollars are more difficult to predict, as we've got two opposing forces at play: (a) upscale cameras being produced to retrieve margin; and (b) existing cameras being discounted to produce volume and clear out things like sensor commitments.
My take on the Nikkei article is this: the company with half the consumer camera market doesn't think we're out of the decline period yet. And I 100% agree. I see no signal that anything really has changed, other than perhaps a faster shift from full frame DSLR to mirrorless than some imagined. Which in turn will likely trigger Canon and Nikon to do more to make the same transition in crop-sensor.
Back in 2012, when it was clear that we'd reached peak ILC, I was interviewed on national radio and asked about how far the volume would fall. My answer was three-pronged: (1) yes, it would fall to an unknown floor; (2) the floor could be as low as late film SLR (roughly 6m+ units a year) factored for population and global growth; and (3) was likely to be around 8m units as far as I could tell. (As a reminder, I predicted in 2003 that we would hit peak ILC in 2011, and was off by perhaps about six months.)
I haven't really changed my answer since 2012. We're down to 10.8m units (from 16.2m in 2012). A decline to 8m units is another 26% decline, and that's not difficult to imagine now. Think back to my points about things I don't consider boding strongly for ILC in the next few years: (1) as more people transition from DSLR to mirrorless, there are fewer people sampling over and over to see if "it's ready yet"; (2) that hangover in the inventory chain can't persist forever (e.g. an original Sony A7 isn't really viable much longer at any price); and (3) the closer we get to bottom, the more the manufacturers want to hold or improve their gross profit margin.
So what does this all mean for the ILC camera buyer?
- Fewer ILC products. An 8m unit market can't support 50+ models. We're going to see some model lines go away. Realistically, you need an entry consumer, high consumer, entry prosumer, high prosumer, and a pro product. The days of putting more products in the queue to finesse out a few more marginal sales are long gone. The volume isn't there to support that, and the R&D and sales costs of doing that are too high.
- Slower updates. The top models have tended to be on 24-month update cycles. If I were in charge, I'd be looking at 36-month or 48-month update cycles with a mid-term large firmware update. You can't lengthen the cycles too much, or else you risk lowering the volume more, but you want to lengthen the hardware side of the cycle so you're not pushing too much cash each year into physical R&D. You keep the existing model in people's minds by making sure it moves forward with features and performance through firmware updates.
- Pricing has to stabilize higher. The problem that Sony made for the market is that they established a <US$1000 full frame camera (original A7 in recent sales). That's super dangerous. If the expectation is that entry full frame is US$1000, that's going to make it difficult to sell cameras profitably in the future. It's really tough to cut costs at the sensor, and the sensor is usually the highest cost component in the camera. If we want better sensors in the future, there has to be cashflow to the camera companies that allows R&D investment. US$1000 full frame cameras are not going to provide that.
- The transition is here. Mirrorless is the future of ILC, though DSLRs will still be around for the foreseeable future. But the fact that the transition is now fully here means something important: old lens mounts are going to see little new activity. If DSLR owners are going to transition to mirrorless (and new buyers start with mirrorless), the lens sale activity is all going to shift to mirrorless. So think RF, L, Z, and FE for the future. Most, if not nearly all, of future lenses are headed for those chunks of metal. (This is one of the reasons why I'm so critical of Canon not rationalizing the EOS M and EOS R systems. That was a product management mistake.)
Thing is, there's not an ILC you can buy today that takes bad photos. That's one of the reasons why this is a slow transition, not a fast one from DSLR to mirrorless. 8-year old DSLRs are still highly viable for most types of shooting (in the Nikon world, that would be a D4 and D800, by the way). So there's no urgency for someone owning one of those products to make the transition; you have to coax them to make the transition by dangling new features and performance. Plus a lot of the ILC crowd is older. While they wouldn't mind lightening their load (smaller, lighter mirrorless over their bigger, heavier DSLR), they also aren't a big fan of change. They don't want to learn something new.
So, my prediction is this: Canon and Nikon will continue to roll a few DSLRs out, though we'll see models start to combine (or models disappear). We won't see many new DSLR lenses, though Canon is more likely at this than Nikon given that their video cameras use the EF mount still. Price will likely start to be used to eek out as much remaining DSLR volume as possible. The recent discount on the D850 is just the beginning, I think.
So don't get caught up in the Internet rhetoric over Canon's statement. The ILC world hasn't changed from what it was before the statement, isn't changing because of the statement. We're in a transition that had already begun, and perhaps now will accelerate a bit.
